CRAN Package Check Results for Package KATforDCEMRI

Last updated on 2020-02-19 10:48:57 CET.

Check Details

Version: 1.0.1
Check: examples
Result: ERROR
    Running examples in 'KATforDCEMRI-Ex.R' failed
    The error most likely occurred in:
    
    > base::assign(".ptime", proc.time(), pos = "CheckExEnv")
    > ### Name: KAT
    > ### Title: Kinetic Analysis Tool for DCE-MRI
    > ### Aliases: KAT
    > ### Keywords: kinetic DCEMRI
    >
    > ### ** Examples
    >
    > ## Create temporary directory for example code output files
    > temp_dir <- tempdir(check=FALSE)
    > ##
    > current_dir <- getwd()
    > setwd(temp_dir)
    > ##
    > ## Run example code
    > demo(KAT, ask=FALSE)
    
    
     demo(KAT)
     ---- ~~~
    
    > ## KATforDCEMRI: a Kinetic Analysis Tool for DCE-MRI
    > ## Copyright 2018 Genentech, Inc.
    > ##
    > ## For questions or comments, please contact
    > ## Gregory Z. Ferl, Ph.D.
    > ## Genentech, Inc.
    > ## Development Sciences
    > ## 1 DNA Way, Mail stop 463A
    > ## South San Francisco, CA, United States of America
    > ## E-mail: ferl.gregory@gene.com
    >
    > runme <- function(){
    + data(dcemri.data, package="KATforDCEMRI")
    +
    + ## dir.create("KATforDCEMRI_benchmark_test")
    + ## setwd("KATforDCEMRI_benchmark_test")
    +
    + attach(dcemri.data)
    +
    + ## SHRINK THE ROI MASK
    + maskROI[,,] <- 0
    + #maskROI[32:42,32:42,] <- 1
    + maskROI[34:36,34:36,] <- 1
    +
    + runtime1 <- system.time(KAT.checkData(file.name="KAT", vector.times=vectorTimes, map.CC=mapCC, mask.ROI=maskROI, vector.AIF=vectorAIF))
    + runtime2 <- system.time(KAT(file = "KAT.RData", results_file="KAT_benchmark_test-full", range.map=1.05, cutoff.map=0.95, AIF.shift="NONE", tlag.Tofts.on=FALSE, export.matlab=FALSE))
    +
    + ## runtime3 <- system.time(KAT.checkData(file.name="KATtrunc", vector.times=vectorTimes[1:44], map.CC=mapCC[,,,1:44], mask.ROI=maskROI, vector.AIF=vectorAIF[1:44]))
    + ## runtime4 <- system.time(KAT(file = "KATtrunc.RData", results_file="KAT_benchmark_test-truncated", range.map=1.05, cutoff.map=0.95))
    + detach(dcemri.data)
    +
    + ## runtime <- format(runtime1[3] + runtime2[3] + runtime3[3] + runtime4[3], digits=3)
    + runtime <- format(runtime1[3] + runtime2[3], digits=3)
    +
    + ## KAT.plot(F1="KAT_benchmark_test-full_slice1.RData", F2="KAT_benchmark_test-full_slice2.RData", F3="KAT_benchmark_test-truncated_slice1.RData", F4="KAT_benchmark_test-truncated_slice2.RData", export.matlab=FALSE)
    + KAT.plot(F1="KAT_benchmark_test-full_slice1.RData", F2="KAT_benchmark_test-full_slice2.RData", F3="KAT_benchmark_test-full_slice3.RData", F4="KAT_benchmark_test-full_slice4.RData", export.matlab=FALSE)
    +
    + load("KAT_benchmark_test-full_slice1.RData")
    + Ktrans_s1A <- as.numeric(format(dcemri.data$paramestmedianxT$Ktrans.median, digits=3))
    + kep_s1A <- as.numeric(format(dcemri.data$paramestmedianxT$kep.median, digits=3))
    + vb_s1A <- as.numeric(format(dcemri.data$paramestmedianxT$vb.median, digits=3))
    + cvKtrans_s1A <- as.numeric(format(100*sd(as.vector(dcemri.data$mapKtransxT), na.rm=TRUE)/Ktrans_s1A, digits=3))
    + cvkep_s1A <- as.numeric(format(100*sd(as.vector(dcemri.data$mapkepxT), na.rm=TRUE)/kep_s1A, digits=3))
    + cvvb_s1A <- as.numeric(format(100*sd(as.vector(dcemri.data$mapvbxT), na.rm=TRUE)/vb_s1A, digits=3))
    +
    + Ktrans_Ts1A <- as.numeric(format(dcemri.data$paramestmedianT$Ktrans.median, digits=3))
    + kep_Ts1A <- as.numeric(format(dcemri.data$paramestmedianT$kep.median, digits=3))
    + cvKtrans_Ts1A <- as.numeric(format(100*sd(as.vector(dcemri.data$mapKtransT), na.rm=TRUE)/Ktrans_Ts1A, digits=3))
    + cvkep_Ts1A <- as.numeric(format(100*sd(as.vector(dcemri.data$mapkepT), na.rm=TRUE)/kep_s1A, digits=3))
    +
    + load("KAT_benchmark_test-full_slice2.RData")
    + Ktrans_s2A <- as.numeric(format(dcemri.data$paramestmedianxT$Ktrans.median, digits=3))
    + kep_s2A <- as.numeric(format(dcemri.data$paramestmedianxT$kep.median, digits=3))
    + vb_s2A <- as.numeric(format(dcemri.data$paramestmedianxT$vb.median, digits=3))
    + cvKtrans_s2A <- as.numeric(format(100*sd(as.vector(dcemri.data$mapKtransxT), na.rm=TRUE)/Ktrans_s2A, digits=3))
    + cvkep_s2A <- as.numeric(format(100*sd(as.vector(dcemri.data$mapkepxT), na.rm=TRUE)/kep_s2A, digits=3))
    + cvvb_s2A <- as.numeric(format(100*sd(as.vector(dcemri.data$mapvbxT), na.rm=TRUE)/vb_s2A, digits=3))
    +
    + Ktrans_Ts2A <- as.numeric(format(dcemri.data$paramestmedianT$Ktrans.median, digits=3))
    + kep_Ts2A <- as.numeric(format(dcemri.data$paramestmedianT$kep.median, digits=3))
    + cvKtrans_Ts2A <- as.numeric(format(100*sd(as.vector(dcemri.data$mapKtransT), na.rm=TRUE)/Ktrans_Ts2A, digits=3))
    + cvkep_Ts2A <- as.numeric(format(100*sd(as.vector(dcemri.data$mapkepT), na.rm=TRUE)/kep_s2A, digits=3))
    +
    + ## load("KAT_benchmark_test-truncated_slice1.RData")
    + load("KAT_benchmark_test-full_slice3.RData")
    + Ktrans_s1B <- as.numeric(format(dcemri.data$paramestmedianxT$Ktrans.median, digits=3))
    + kep_s1B <- as.numeric(format(dcemri.data$paramestmedianxT$kep.median, digits=3))
    + vb_s1B <- as.numeric(format(dcemri.data$paramestmedianxT$vb.median, digits=3))
    + cvKtrans_s1B <- as.numeric(format(100*sd(as.vector(dcemri.data$mapKtransxT), na.rm=TRUE)/Ktrans_s1B, digits=3))
    + cvkep_s1B <- as.numeric(format(100*sd(as.vector(dcemri.data$mapkepxT), na.rm=TRUE)/kep_s1B, digits=3))
    + cvvb_s1B <- as.numeric(format(100*sd(as.vector(dcemri.data$mapvbxT), na.rm=TRUE)/vb_s1B, digits=3))
    +
    + Ktrans_Ts1B <- as.numeric(format(dcemri.data$paramestmedianT$Ktrans.median, digits=3))
    + kep_Ts1B <- as.numeric(format(dcemri.data$paramestmedianT$kep.median, digits=3))
    + cvKtrans_Ts1B <- as.numeric(format(100*sd(as.vector(dcemri.data$mapKtransT), na.rm=TRUE)/Ktrans_Ts1B, digits=3))
    + cvkep_Ts1B <- as.numeric(format(100*sd(as.vector(dcemri.data$mapkepT), na.rm=TRUE)/kep_s1B, digits=3))
    +
    + ## load("KAT_benchmark_test-truncated_slice2.RData")
    + load("KAT_benchmark_test-full_slice4.RData")
    + Ktrans_s2B <- as.numeric(format(dcemri.data$paramestmedianxT$Ktrans.median, digits=3))
    + kep_s2B <- as.numeric(format(dcemri.data$paramestmedianxT$kep.median, digits=3))
    + vb_s2B <- as.numeric(format(dcemri.data$paramestmedianxT$vb.median, digits=3))
    + cvKtrans_s2B <- as.numeric(format(100*sd(as.vector(dcemri.data$mapKtransxT), na.rm=TRUE)/Ktrans_s2B, digits=3))
    + cvkep_s2B <- as.numeric(format(100*sd(as.vector(dcemri.data$mapkepxT), na.rm=TRUE)/kep_s2B, digits=3))
    + cvvb_s2B <- as.numeric(format(100*sd(as.vector(dcemri.data$mapvbxT), na.rm=TRUE)/vb_s2B, digits=3))
    +
    + Ktrans_Ts2B <- as.numeric(format(dcemri.data$paramestmedianT$Ktrans.median, digits=3))
    + kep_Ts2B <- as.numeric(format(dcemri.data$paramestmedianT$kep.median, digits=3))
    + cvKtrans_Ts2B <- as.numeric(format(100*sd(as.vector(dcemri.data$mapKtransT), na.rm=TRUE)/Ktrans_Ts2B, digits=3))
    + cvkep_Ts2B <- as.numeric(format(100*sd(as.vector(dcemri.data$mapkepT), na.rm=TRUE)/kep_s2B, digits=3))
    +
    + pdf(file="KAT_demo-page1.pdf", height=11, width=8.5)
    +
    + plot(0:35, type = "n", xaxt="n", yaxt="n", bty="n", xlab = "", ylab = "")
    + text(5, 35, paste("KATforDCEMRI version ", dcemri.data$KATversion, " BENCHMARK TEST", sep=""), pos=4, font=2, col="red")
    + text(5, 33, paste("date:", date(), "\n"), pos=4)
    + text(5, 32, paste("Total Processing Time:", runtime, "seconds"), pos=4)
    + text(5, 29, paste("sysname/release:", Sys.info()[[1]], Sys.info()[[2]], "\n"), pos=4)
    + text(5, 28, paste("nodename:", Sys.info()[[4]], "\n"), pos=4)
    + text(5, 27, paste("user:", Sys.info()[[7]], "\n"), pos=4)
    + text(5, 25, "Estimated (inter-voxel %CV) versus True Parameter values (extended Tofts)", pos=4, font=2)
    + text(5, 23, "LOW NOISE DATA", pos=4)
    + text(5, 22, paste("Slice 1: Ktrans =", Ktrans_s1A, "1/min (", cvKtrans_s1A,"%) [true value = 0.22 1/min]", sep=""), pos=4)
    + text(5, 21, paste("Slice 1: kep =", kep_s1A, "1/min (", cvkep_s1A, "%) [true value = 1.1 1/min]", sep=""), pos=4)
    + text(5, 20, paste("Slice 1: vb =", vb_s1A, " (", cvvb_s1A, "%) [true value = 0]", sep=""), pos=4)
    + text(5, 18, paste("Slice 2: Ktrans =", Ktrans_s2A, "1/min (", cvKtrans_s2A,"%) [true value = 0.22 1/min]", sep=""), pos=4)
    + text(5, 17, paste("Slice 2: kep =", kep_s2A, "1/min (", cvkep_s2A, "%) [true value = 1.1 1/min]", sep=""), pos=4)
    + text(5, 16, paste("Slice 2: vb =", vb_s2A, " (", cvvb_s2A, "%) [true value = 0.05]", sep=""), pos=4)
    + text(5, 14, "HIGH NOISE DATA", pos=4)
    + text(5, 13, paste("Slice 3: Ktrans =", Ktrans_s1B, "1/min (", cvKtrans_s1B,"%) [true value = 0.22 1/min]", sep=""), pos=4)
    + text(5, 12, paste("Slice 3: kep =", kep_s1B, "1/min (", cvkep_s1B, "%) [true value = 1.1 1/min]", sep=""), pos=4)
    + text(5, 11, paste("Slice 3: vb =", vb_s1B, " (", cvvb_s1B, "%) [true value = 0]", sep=""), pos=4)
    + text(5, 9, paste("Slice 4: Ktrans =", Ktrans_s2B, "1/min (", cvKtrans_s2B,"%) [true value = 0.22 1/min]", sep=""), pos=4)
    + text(5, 8, paste("Slice 4: kep =", kep_s2B, "1/min (", cvkep_s2B, "%) [true value = 1.1 1/min]", sep=""), pos=4)
    + text(5, 7, paste("Slice 4: vb =", vb_s2B, " (", cvvb_s2B, "%) [true value = 0.05]", sep=""), pos=4)
    + dev.off()
    +
    +
    + pdf(file="KAT_demo-page2.pdf", height=11, width=8.5)
    +
    + plot(0:35, type = "n", xaxt="n", yaxt="n", bty="n", xlab = "", ylab = "")
    +
    + text(5, 25, "Estimated (inter-voxel %CV) versus True Parameter values (Tofts Model)", pos=4, font=2)
    + text(5, 23, "LOW NOISE DATA", pos=4)
    + text(5, 22, paste("Slice 1: Ktrans =", Ktrans_Ts1A, "1/min (", cvKtrans_Ts1A,"%) [true value = 0.22 1/min]", sep=""), pos=4)
    + text(5, 21, paste("Slice 1: kep =", kep_Ts1A, "1/min (", cvkep_Ts1A, "%) [true value = 1.1 1/min]", sep=""), pos=4)
    + text(5, 20, "Slice 1: vb = 0 (Fixed) [true value = 0]", pos=4)
    + text(5, 18, paste("Slice 2: Ktrans =", Ktrans_Ts2A, "1/min (", cvKtrans_Ts2A,"%) [true value = 0.22 1/min]", sep=""), pos=4)
    + text(5, 17, paste("Slice 2: kep =", kep_Ts2A, "1/min (", cvkep_Ts2A, "%) [true value = 1.1 1/min]", sep=""), pos=4)
    + text(5, 16, "Slice 2: vb = 0 (Fixed) [true value = 0.05]", pos=4)
    + text(5, 14, "HIGH NOISE DATA", pos=4)
    + text(5, 13, paste("Slice 3: Ktrans =", Ktrans_Ts1B, "1/min (", cvKtrans_Ts1B,"%) [true value = 0.22 1/min]", sep=""), pos=4)
    + text(5, 12, paste("Slice 3: kep =", kep_Ts1B, "1/min (", cvkep_Ts1B, "%) [true value = 1.1 1/min]", sep=""), pos=4)
    + text(5, 11, "Slice 3: vb = 0 (Fixed) [true value = 0]", pos=4)
    + text(5, 9, paste("Slice 4: Ktrans =", Ktrans_Ts2B, "1/min (", cvKtrans_Ts2B,"%) [true value = 0.22 1/min]", sep=""), pos=4)
    + text(5, 8, paste("Slice 4: kep =", kep_Ts2B, "1/min (", cvkep_Ts2B, "%) [true value = 1.1 1/min]", sep=""), pos=4)
    + text(5, 7, "Slice 4: vb = 0 (Fixed) [true value = 0.05]", pos=4)
    + dev.off()
    + }
    
    > runme()
    
    checking dimensions of vectors and arrays...
    
    length of vector.times is 89 with units of seconds
    length of vector.AIF is 89
    dimensions of map.CC array are 75 x 75 x 4 slices x 89 time points
    dimensions of mask.ROI array are 75 x 75 x 4 slices
    
    ...vector and array dimensions are okay.
    
    Saving data in a single R file...
    ...file saved as KAT.RData ...
    ...use the KAT() function to analyze data within this file.
    
    
    #########################################################################
    ##-------------- KINETIC ANALYSIS TOOL (KAT) FOR DCE-MRI --------------##
    #########################################################################
    ##---------------------- R package version 1.0 ----------------------##
    #########################################################################
    
    loading KAT.RData into R...
    ..done in 0.0012 minutes.
    --------
    ***** ROI DETECTED IN SLICE 1 *****
    --------
    extracting slice 1 for analysis...
    ..done in 0.0018 minutes.
    --------
    applying ROI mask to cc matrix...
    ..done in 0.00032 minutes.
    --------
    fitting xTofts and Tofts models to whole ROI data...
    ..done in 0.0033 minutes.
    --------
     ----------- FAILURE REPORT --------------
     --- failure: the condition has length > 1 ---
     --- srcref ---
    :
     --- package (from environment) ---
    KATforDCEMRI
     --- call from context ---
    KAT(file = "KAT.RData", results_file = "KAT_benchmark_test-full",
     range.map = 1.05, cutoff.map = 0.95, AIF.shift = "NONE",
     tlag.Tofts.on = FALSE, export.matlab = FALSE)
     --- call from argument ---
    if (class(cov) != "try-error") {
     sd <- try(sqrt(diag(cov)), silent = try.silent)
     if (class(sd) != "try-error") {
     cv <- sd/param_est * 100
     cv.whole.roi.xTofts <- as.numeric(format(cv, digits = 1))
     }
    }
     --- R stacktrace ---
    where 1 at /home/hornik/tmp/R.check/r-devel-clang/Work/build/Packages/KATforDCEMRI/demo/KAT.R#26: KAT(file = "KAT.RData", results_file = "KAT_benchmark_test-full",
     range.map = 1.05, cutoff.map = 0.95, AIF.shift = "NONE",
     tlag.Tofts.on = FALSE, export.matlab = FALSE)
    where 2 at /home/hornik/tmp/R.check/r-devel-clang/Work/build/Packages/KATforDCEMRI/demo/KAT.R#26: system.time(KAT(file = "KAT.RData", results_file = "KAT_benchmark_test-full",
     range.map = 1.05, cutoff.map = 0.95, AIF.shift = "NONE",
     tlag.Tofts.on = FALSE, export.matlab = FALSE))
    where 3 at /home/hornik/tmp/R.check/r-devel-clang/Work/build/Packages/KATforDCEMRI/demo/KAT.R#141: runme()
    where 4: eval(ei, envir)
    where 5: eval(ei, envir)
    where 6: withVisible(eval(ei, envir))
    where 7: source(available, echo = echo, max.deparse.length = Inf, keep.source = TRUE,
     encoding = encoding)
    where 8: demo(KAT, ask = FALSE)
    
     --- value of length: 2 type: logical ---
    [1] TRUE TRUE
     --- function from context ---
    function (file = "concatenate.KAT.with.KAT.checkData.RData",
     results_file = "my_results", method.optimization = "L-BFGS-B",
     show.rt.fits = FALSE, param.for.avdt = "Ktrans", range.map = 1.5,
     cutoff.map = 0.85, export.matlab = TRUE, export.RData = TRUE,
     verbose = FALSE, show.errors = FALSE, try.silent = TRUE,
     fracGTzero = 0.75, AIF.shift = "", Force.AIF.peak = FALSE,
     tlag.Tofts.on = FALSE, est.per.voxel.tlag = FALSE, ...)
    {
     lo <- 0
     options(show.error.messsages = show.errors)
     ftype <- strsplit(file, split = "a")[[1]]
     ftype <- ftype[length(ftype) - 1]
     ftype <- strsplit(ftype, split = "")[[1]]
     ftype <- ftype[length(ftype)]
     if (ftype == "m")
     file.format <- "matlab"
     if (ftype == "D")
     file.format <- "RData"
     file_short <- file
     KAT.version <- "1.0"
     ptm_total <- proc.time()[3]
     modeltype1 <- "xTofts"
     modeltype2 <- "Tofts"
     funcz <- function(x) {
     as.numeric(as.character(x))
     }
     aif.shift.func <- function(t, cp, time_shift) {
     x <- cbind(t, cp)
     tmax <- subset(x[, 1], x[, 2] == max(x[, 2]))
     if (AIF.shift == "ARTERY")
     tmax <- tmax + time_shift
     if (AIF.shift == "VEIN")
     tmax <- tmax - time_shift
     cpFUNC <- approxfun(t, cp, rule = 2)
     if (AIF.shift == "ARTERY")
     tshift <- t - time_shift
     if (AIF.shift == "VEIN")
     tshift <- t + time_shift
     cp.shift <- cpFUNC(tshift)
     if (Force.AIF.peak == TRUE)
     cp.shift[cp.shift == max(cp.shift)] <- max(cp)
     return(cp.shift)
     }
     roi.modelT <- function(p, t, dt, cp, zing = 0) {
     if (zing == 0) {
     Ktrans <- p[1]
     kep <- p[2]
     if (tlag.Tofts.on == TRUE) {
     if (fix.tlag != TRUE & AIF.shift != "NONE")
     time_shift <- p[3]
     if (fix.tlag == TRUE & AIF.shift != "NONE")
     time_shift <- param.est.whole.roi.Tofts$tlag
     }
     if (tlag.Tofts.on == FALSE) {
     time_shift <- param.est.whole.roi.xTofts$tlag
     }
     if (AIF.shift == "ARTERY" | AIF.shift == "VEIN")
     cp <- aif.shift.func(t, cp, time_shift)
     f <- Ktrans * exp(kep * t) * cp
     int <- c(0, cumsum(dt * (tail(f, -1) + head(f, -1)))/2)
     ct <- exp(-kep * t) * int
     return(ct)
     }
     if (zing == 1)
     return("modelT")
     }
     roi.modelxT <- function(p, t, dt, cp, zing = 0) {
     if (zing == 0) {
     Ktrans <- p[1]
     kep <- p[2]
     vb <- p[3]
     if (fix.tlag != TRUE & AIF.shift != "NONE")
     time_shift <- p[4]
     if (est.per.voxel.tlag == FALSE) {
     if (fix.tlag == TRUE & AIF.shift != "NONE")
     time_shift <- param.est.whole.roi.xTofts$tlag
     }
     if (est.per.voxel.tlag == TRUE) {
     if (fix.tlag == TRUE & AIF.shift != "NONE")
     time_shift <- p[4]
     }
     if (AIF.shift == "ARTERY" | AIF.shift == "VEIN")
     cp <- aif.shift.func(t, cp, time_shift)
     f <- Ktrans * exp(kep * t) * cp
     int <- c(0, cumsum(dt * (tail(f, -1) + head(f, -1)))/2)
     ct <- exp(-kep * t) * int
     ct <- ct + vb * cp
     return(ct)
     }
     if (zing == 1)
     return("modelxT")
     }
     calch <- function(u, y, TIME_trunc) {
     locfit_y <- preplot(y, newdata = 0:max(TIME_trunc))
     y_smooth <- locfit_y$fit
     u <- u[match(u[u == max(u)], u):length(u)]
     y_smooth <- y_smooth[match(u[u == max(u)], u):length(u)]
     n <- length(u)
     A <- matrix(0, nrow = n, ncol = n)
     ind <- row(A) - col(A)
     ind[ind < 0] <- (-1)
     ind <- ind + 2
     A <- matrix(c(0, u)[ind], nrow = n, ncol = n)
     h <- solve(A, y_smooth)
     h.time.vector <- (1:length(h)) - 1
     out <- list(h.time.vector, h)
     names(out) <- c("t", "IRF")
     return(out)
     }
     calchFUNC <- function(vector.times, AIF, map_cc_slice, correct.vp = TRUE,
     alpha.AIF = c(0.1, 0.5), vp.nom = 0.1, kep.nom = 0.5) {
     AUMC <- function(AUMC.median, h.median, irf_time_vec,
     r) {
     AUMC.median <- AUMC.median + 0.5 * (h.median[r] *
     irf_time_vec[r] + h.median[r + 1] * irf_time_vec[r +
     1])
     }
     artery_data <- data.frame(vector.times * 60, AIF)
     names(artery_data) <- c("TIME", "ARTERY")
     data_artery_peak <- subset(artery_data, artery_data$ARTERY ==
     max(artery_data$ARTERY))
     data_remove_artery_prepeak <- subset(artery_data, artery_data$TIME >=
     data_artery_peak$TIME)
     frames_to_peak <- length(artery_data[, 1]) - length(data_remove_artery_prepeak[,
     1]) + 1
     TIME <- data_remove_artery_prepeak$TIME
     ARTERY <- data_remove_artery_prepeak$ARTERY
     TIME_trunc <- TIME[seq(1, length(TIME) - 1, by = 1)]
     TIME_trunc <- TIME_trunc - TIME_trunc[1]
     ARTERY_trunc <- ARTERY[seq(1, length(ARTERY) - 1, by = 1)]
     ARTERY_smooth <- locfit.robust(ARTERY_trunc ~ TIME_trunc,
     acri = "cp", alpha = alpha.AIF)
     AIF_smooth <- ARTERY_smooth
     locfit_u <- preplot(AIF_smooth, newdata = 0:max(TIME_trunc))
     u_smooth <- locfit_u$fit
     Tmax <- max(TIME_trunc)
     TUMOR.median <- map_cc_slice
     TUMOR.median <- TUMOR.median[seq(frames_to_peak, length(TUMOR.median),
     by = 1)]
     if (vp.nom > 0)
     TUMOR.median_corr <- TUMOR.median - vp.nom * ARTERY
     if (vp.nom <= 0)
     TUMOR.median_corr <- TUMOR.median
     TUMOR.median_corr_shifted <- TUMOR.median_corr[seq(2,
     length(TUMOR.median_corr), by = 1)]
     TUMOR.median_smooth <- locfit.robust(TUMOR.median_corr_shifted ~
     TIME_trunc, acri = "cp")
     calch.out <- calch(u_smooth, TUMOR.median_smooth, TIME_trunc)
     h.median <- calch.out$IRF
     irf_time_vec <- calch.out$t
     n <- length(h.median)
     AUC.median <- 0
     AUMC.median <- 0
     for (r in 1:(n - 1)) {
     h_sum <- h.median[r] + h.median[r + 1]
     t_sum <- irf_time_vec[r] + irf_time_vec[r + 1]
     AUC.median <- AUC.median + 0.5 * h_sum
     AUMC.median <- AUMC(AUMC.median, h.median, irf_time_vec,
     r)
     }
     AUCMRT.median <- AUC.median/(AUMC.median/AUC.median) *
     60
     if (kep.nom > 0) {
     t_scan <- max(TIME_trunc)/60
     ve_trunc_error <- 1 - exp(-kep.nom * t_scan)
     Ktrans_trunc_error <- (1 - exp(-kep.nom * t_scan))^2/(1 -
     (1 + kep.nom * t_scan) * exp(-kep.nom * t_scan))
     AUC.median <- AUC.median/ve_trunc_error
     AUCMRT.median <- AUCMRT.median/Ktrans_trunc_error
     }
     irf_time_vec <- irf_time_vec/60
     out <- list(AUC.median, AUCMRT.median, h.median, irf_time_vec)
     names(out) <- c("AUCh", "AUChMRTh", "IRF", "t")
     return(out)
     }
     Obj_roi <- function(p, model, t, dt, cp, roi) {
     sum((model(p, t, dt, cp) - roi)^2)
     }
     map_cc_slice <- NULL
     map_cc_roi <- NULL
     aif <- NULL
     aif.shifted <- NULL
     map.times <- NULL
     map.KtransxT <- NULL
     map.kepxT <- NULL
     map.vexT <- NULL
     map.vbxT <- NULL
     map.tlagxT <- NULL
     map.fitfailuresxT <- NULL
     map.KtransT <- NULL
     map.kepT <- NULL
     map.veT <- NULL
     map.fitfailuresT <- NULL
     mask.roi <- NULL
     nx <- NULL
     ny <- NULL
     nt <- NULL
     map.KtransxT <- NULL
     map.kepxT <- NULL
     map.vexT <- NULL
     map.vbxT <- NULL
     map.AIC.compare <- NULL
     map.AIC.T <- NULL
     map.EF <- NULL
     map.AIC.xT <- NULL
     roi.median.fitted.Tofts <- NULL
     param.est.whole.roi.Tofts <- NULL
     roi.median.fitted.xTofts <- NULL
     param.est.whole.roi.xTofts <- NULL
     cv.whole.roi.xTofts <- NULL
     cat("\n")
     cat("#########################################################################",
     "\n")
     cat("##-------------- KINETIC ANALYSIS TOOL (KAT) FOR DCE-MRI --------------##",
     "\n")
     cat("#########################################################################",
     "\n")
     cat("##---------------------- R package version", KAT.version,
     "----------------------##", "\n")
     cat("#########################################################################",
     "\n")
     cat("\n")
     filea <- strsplit(file, split = "/")[[1]]
     fileb <- filea[length(filea)]
     if (file != "concatenate.KAT.with.KAT.checkData.RData")
     cat("loading", fileb, "into R...", "\n")
     ptm <- proc.time()[3]
     if (file.format == "matlab") {
     mat_data <- readMat(file)
     if (length(names(mat_data)) < 10)
     file.original <- TRUE
     if (length(names(mat_data)) >= 10)
     file.original <- FALSE
     }
     if (file.format == "RData") {
     if (file != "concatenate.KAT.with.KAT.checkData.RData")
     load(file)
     if (length(names(dcemri.data)) < 10) {
     file.original <- TRUE
     ROIcounter <- apply(dcemri.data$maskROI, 3, max)
     results_file_temp <- results_file
     }
     if (length(names(dcemri.data)) >= 10) {
     file.original <- FALSE
     ROIcounter <- 1
     }
     }
     if (file != "concatenate.KAT.with.KAT.checkData.RData") {
     cat("..done in", format((proc.time()[3] - ptm)/60, digits = 2),
     "minutes.", "\n")
     cat("--------", "\n")
     }
     for (slicenumber in 1:(length(ROIcounter))) {
     if (ROIcounter[slicenumber] == 1) {
     if (file.original == TRUE) {
     slice <- slicenumber
     cat("***** ROI DETECTED IN SLICE", slice, " *****",
     "\n")
     cat("--------", "\n")
     }
     if (file.original == TRUE) {
     if (AIF.shift != "VEIN" & AIF.shift != "ARTERY" &
     AIF.shift != "NONE")
     stop("You must specify the argument AIF.shift argument as VEIN, ARTERY or NONE, indicating that the AIF you are using is based on data from a vein or artery or NONE if tlag should be set to 0. This will ensure that the time lag parameter in the Tofts and xTofts models has the appropriate inital value and is bounded correctly; either -Inf to 0 (for VEIN) or 0 to Inf (for ARTERY).")
     filenameTag <- paste("_slice", slice, sep = "")
     results_file <- paste(results_file_temp, filenameTag,
     sep = "")
     roi.model <- roi.modelxT
     if (slice == "" || slice < 0)
     stop("The slice argument has not been properly specified; slice=``slice number''")
     cat("extracting slice", slice, "for analysis...",
     "\n")
     ptm <- proc.time()[3]
     if (file.format == "matlab") {
     map.times <- as.vector(mat_data$map[[4]]/60)
     map_cc <- mat_data$map[[3]]
     map_cc_slice <- map_cc[, , slice, ]
     if (length(map.times) != length(map_cc_slice[1,
     1, ]))
     stop("The length of the time vector does not match the length of the contrast agent concentration vector")
     }
     if (file.format == "RData") {
     map.times <- as.vector(dcemri.data$vectorTimes/60)
     map_cc <- dcemri.data$mapCC
     map_cc_slice <- map_cc[, , slice, ]
     if (length(map.times) != length(map_cc_slice[1,
     1, ]))
     stop("The length of the time vector does not match the length of the contrast agent concentration vector")
     }
     nt <- length(map_cc_slice[1, 1, ])
     ny <- length(map_cc_slice[, 1, 1])
     nx <- length(map_cc_slice[1, , 1])
     ccTEMP <- rep(0, prod(dim(map_cc_slice)))
     dim(ccTEMP) <- dim(map_cc_slice)[c(2, 1, 3)]
     for (i in 1:nt) ccTEMP[, , i] <- rot90(map_cc_slice[,
     , i], 3)
     map_cc_slice <- ccTEMP
     if (file.format == "matlab")
     mask.roi <- mat_data$mask[[1]][, , slice]
     if (file.format == "RData")
     mask.roi <- dcemri.data$mask[, , slice]
     mask.roi <- rot90(mask.roi, 3)
     if (max(mask.roi) != 1)
     stop("Your ROI mask is either composed entirely of zeroes or contains nonnumeric elements; voxels within the ROI should have a value of ``1'' and all other voxels should have a value of ``0''.")
     if (file.format == "matlab")
     aif <- as.vector(mat_data$aif)
     if (file.format == "RData")
     aif <- as.vector(dcemri.data$vectorAIF)
     if (file.format == "matlab")
     rm(mat_data)
     cat("..done in", format((proc.time()[3] - ptm)/60,
     digits = 2), "minutes.", "\n")
     cat("--------", "\n")
     cat("applying ROI mask to cc matrix...", "\n")
     ptm <- proc.time()[3]
     map_cc_roi <- map_cc_slice
     mask.roi.temp <- mask.roi
     mask.roi.temp[mask.roi.temp == 0] <- NA
     for (z in 1:nt) map_cc_roi[, , z] <- map_cc_roi[,
     , z] * mask.roi.temp
     cat("..done in", format((proc.time()[3] - ptm)/60,
     digits = 2), "minutes.", "\n")
     cat("--------", "\n")
     ptm <- proc.time()[3]
     cc.median <- seq(1:nt)
     for (t in 1:nt) cc.median[t] <- median(map_cc_roi[,
     , t], na.rm = TRUE)
     cat("fitting", modeltype1, "and", modeltype2,
     "models to whole ROI data...", "\n")
     IRF.out <- calchFUNC(map.times, aif, cc.median)
     AUCcorrnom <- IRF.out$AUCh
     AUCMRTcorrnom <- IRF.out$AUChMRTh
     AUCMRTcorrnom.divby.AUCcorrnom <- IRF.out$AUChMRTh/IRF.out$AUCh
     if (tlag.Tofts.on == TRUE & AIF.shift == "VEIN") {
     p0.T.median <- c(IRF.out$AUChMRTh, IRF.out$AUChMRTh/IRF.out$AUCh,
     0.05)
     lower.wholeT <- c(lo, lo, 0)
     upper.wholeT <- c(Inf, Inf, Inf)
     }
     if (tlag.Tofts.on == TRUE & AIF.shift == "ARTERY") {
     p0.T.median <- c(IRF.out$AUChMRTh, IRF.out$AUChMRTh/IRF.out$AUCh,
     0.05)
     lower.wholeT <- c(lo, lo, 0)
     upper.wholeT <- c(Inf, Inf, Inf)
     }
     if (tlag.Tofts.on == FALSE) {
     p0.T.median <- c(IRF.out$AUChMRTh, IRF.out$AUChMRTh/IRF.out$AUCh)
     lower.wholeT <- c(lo, lo)
     upper.wholeT <- c(Inf, Inf)
     }
     if (AIF.shift == "VEIN") {
     p0.xT.median <- c(IRF.out$AUChMRTh, IRF.out$AUChMRTh/IRF.out$AUCh,
     0.05, 0.05)
     lower.wholexT <- c(lo, lo, lo, 0)
     upper.wholexT <- c(Inf, Inf, Inf, Inf)
     }
     if (AIF.shift == "ARTERY") {
     p0.xT.median <- c(IRF.out$AUChMRTh, IRF.out$AUChMRTh/IRF.out$AUCh,
     0.05, 0.05)
     lower.wholexT <- c(lo, lo, lo, 0)
     upper.wholexT <- c(Inf, Inf, Inf, Inf)
     }
     if (AIF.shift == "NONE") {
     p0.xT.median <- c(IRF.out$AUChMRTh, IRF.out$AUChMRTh/IRF.out$AUCh,
     0.05)
     lower.wholexT <- c(lo, lo, lo)
     upper.wholexT <- c(Inf, Inf, Inf)
     }
     SAAMII <- FALSE
     if (SAAMII == TRUE) {
     hw.x <- cbind(format(map.times, digits = 3),
     format(aif, digits = 3), format(cc.median,
     digits = 3))
     write.table("DATA", file = paste("slice", slice,
     "_forSAAMII", sep = ""), quote = FALSE, row.names = FALSE,
     col.names = FALSE)
     write.table("(SD 1)", file = paste("slice",
     slice, "_forSAAMII", sep = ""), quote = FALSE,
     row.names = FALSE, append = TRUE, col.names = FALSE)
     write.table(hw.x, file = paste("slice", slice,
     "_forSAAMII", sep = ""), quote = FALSE, row.names = FALSE,
     append = TRUE, col.names = c("t", paste("AIF_s",
     slice, sep = ""), paste("CC_s", slice,
     sep = "")))
     write.table("END", file = paste("slice", slice,
     "_forSAAMII", sep = ""), quote = FALSE, row.names = FALSE,
     append = TRUE, col.names = FALSE)
     }
     roi <- cc.median
     t <- map.times
     fix.tlag <- FALSE
     roi.model <- roi.modelxT
     if (method.optimization == "L-BFGS-B")
     fit.roi.median.xTofts <- try(optim(p0.xT.median,
     Obj_roi, model = roi.model, t = map.times,
     dt = diff(map.times), cp = aif, roi = roi,
     method = "L-BFGS-B", lower = lower.wholexT,
     upper = upper.wholexT, hessian = TRUE), silent = try.silent)
     if (method.optimization != "L-BFGS-B")
     fit.roi.median.xTofts <- try(optim(p0.xT.median,
     Obj_roi, model = roi.model, t = map.times,
     dt = diff(map.times), cp = aif, roi = roi,
     method = method.optimization, hessian = TRUE),
     silent = try.silent)
     if (class(fit.roi.median.xTofts) != "try-error") {
     if (fit.roi.median.xTofts$convergence == 0) {
     roi.median.fitted.xTofts <- roi.model(p = fit.roi.median.xTofts$par,
     t = map.times, dt = diff(map.times), cp = aif)
     params.xTofts <- fit.roi.median.xTofts$par
     param.est.whole.roi.xTofts <- list(fit.roi.median.xTofts$par[1],
     fit.roi.median.xTofts$par[2], fit.roi.median.xTofts$par[3],
     fit.roi.median.xTofts$par[4])
     names(param.est.whole.roi.xTofts) <- c("Ktrans",
     "kep", "vb", "tlag")
     }
     }
     if (class(fit.roi.median.xTofts) == "try-error") {
     cat("A problem occured when fitting the extended Tofts model to median intensity/concentration data across the ROI. Optimization algorithm did not converge. This may occur when a large fraction of voxels within the ROI are non-enhancing, as these are not excluded from analysis of the median intensity/concentration profile",
     "\n")
     }
     if (class(fit.roi.median.xTofts) != "try-error") {
     if (fit.roi.median.xTofts$convergence != 0) {
     cat("A problem occured when fitting the Tofts model to median intensity/concentration data across the ROI (convergence code not equal to zero). Optimization algorithm did not converge. This may occur when a large fraction of voxels within the ROI are non-enhancing, as these are not excluded from analysis of the median intensity/concentration profile",
     "\n")
     }
     }
     roi.model <- roi.modelT
     if (method.optimization == "L-BFGS-B")
     fit.roi.median.Tofts <- try(optim(p0.T.median,
     Obj_roi, model = roi.model, t = map.times,
     dt = diff(map.times), cp = aif, roi = roi,
     method = "L-BFGS-B", lower = lower.wholeT,
     upper = upper.wholeT, hessian = TRUE), silent = try.silent)
     if (method.optimization != "L-BFGS-B")
     fit.roi.median.Tofts <- try(optim(p0.T.median,
     Obj_roi, model = roi.model, t = map.times,
     dt = diff(map.times), cp = aif, roi = roi,
     method = method.optimization, hessian = TRUE),
     silent = try.silent)
     if (class(fit.roi.median.Tofts) != "try-error") {
     if (fit.roi.median.Tofts$convergence == 0) {
     roi.median.fitted.Tofts <- roi.model(p = fit.roi.median.Tofts$par,
     t = map.times, dt = diff(map.times), cp = aif)
     if (tlag.Tofts.on == FALSE) {
     param.est.whole.roi.Tofts <- list(fit.roi.median.Tofts$par[1],
     fit.roi.median.Tofts$par[2])
     names(param.est.whole.roi.Tofts) <- c("Ktrans",
     "kep")
     }
     if (tlag.Tofts.on == TRUE) {
     param.est.whole.roi.Tofts <- list(fit.roi.median.Tofts$par[1],
     fit.roi.median.Tofts$par[2], fit.roi.median.Tofts$par[3])
     names(param.est.whole.roi.Tofts) <- c("Ktrans",
     "kep", "tlag")
     }
     }
     }
     if (class(fit.roi.median.Tofts) == "try-error") {
     cat("A problem occured when fitting the Tofts model to median intensity/concentration data across the ROI (a try-error occured). Optimization algorithm did not converge. This may occur when a large fraction of voxels within the ROI are non-enhancing, as these are not excluded from analysis of the median intensity/concentration profile",
     "\n")
     }
     if (class(fit.roi.median.Tofts) != "try-error") {
     if (fit.roi.median.Tofts$convergence != 0) {
     cat("A problem occured when fitting the Tofts model to median intensity/concentration data across the ROI (convergence code not equal to zero). Optimization algorithm did not converge. This may occur when a large fraction of voxels within the ROI are non-enhancing, as these are not excluded from analysis of the median intensity/concentration profile",
     "\n")
     }
     }
     roi.model <- roi.modelxT
     IRF.out <- calchFUNC(vector.times = map.times,
     AIF = aif, map_cc_slice = cc.median, vp.nom = param.est.whole.roi.xTofts$vb,
     kep.nom = param.est.whole.roi.xTofts$kep)
     p0.T <- c(IRF.out$AUChMRTh, IRF.out$AUChMRTh/IRF.out$AUCh)
     names(p0.T) <- c("Ktrans", "kep")
     if (est.per.voxel.tlag == FALSE) {
     p0.xT <- c(IRF.out$AUChMRTh, IRF.out$AUChMRTh/IRF.out$AUCh,
     param.est.whole.roi.xTofts$vb)
     names(p0.xT) <- c("Ktrans", "kep", "vb")
     }
     if (est.per.voxel.tlag == TRUE) {
     p0.xT <- c(IRF.out$AUChMRTh, IRF.out$AUChMRTh/IRF.out$AUCh,
     param.est.whole.roi.xTofts$vb, 0.05)
     names(p0.xT) <- c("Ktrans", "kep", "vb", "tlag")
     }
     AUCcorr <- IRF.out$AUCh
     AUCMRTcorr <- IRF.out$AUChMRTh
     AUCMRTcorr.divby.AUCcorr <- IRF.out$AUChMRTh/IRF.out$AUCh
     IRF.results <- c(AUCcorrnom, AUCMRTcorrnom, AUCMRTcorrnom.divby.AUCcorrnom,
     AUCcorr, AUCMRTcorr, AUCMRTcorr.divby.AUCcorr)
     names(IRF.results) <- c("AUCcorrnom(ve)", "AUCMRTcorrnom(Ktrans)",
     "AUCMRTcorrnom.divby.AUCcorrnom(kep)", "AUCcorr(ve)",
     "AUCMRTcorr(Ktrans)", "AUCMRTcorr.divby.AUCcorr(kep)")
     cat("..done in", format((proc.time()[3] - ptm)/60,
     digits = 2), "minutes.", "\n")
     cat("--------", "\n")
     roi.model <- roi.modelxT
     if (class(fit.roi.median.xTofts) != "try-error") {
     if (fit.roi.median.xTofts$convergence == 0) {
     RSS <- sum((roi.model(p = fit.roi.median.xTofts$par,
     t = map.times, dt = diff(map.times), cp = aif) -
     roi)^2)
     param_est <- fit.roi.median.xTofts$par
     nD <- nt
     nP <- length(param_est)
     hess <- fit.roi.median.xTofts$hessian
     cov <- try((nP * RSS/(nD - nP)) * solve(hess),
     silent = try.silent)
     cv <- param_est
     cv[] <- NA
     if (class(cov) != "try-error") {
     sd <- try(sqrt(diag(cov)), silent = try.silent)
     if (class(sd) != "try-error") {
     cv <- sd/param_est * 100
     cv.whole.roi.xTofts <- as.numeric(format(cv,
     digits = 1))
     }
     }
     param.roi <- as.numeric(format(param_est,
     digits = 3))
     }
     }
     if (AIF.shift == "ARTERY" | AIF.shift == "VEIN")
     aif.shifted <- aif.shift.func(map.times, aif,
     param.est.whole.roi.xTofts$tlag)
     if (show.rt.fits == TRUE) {
     if (AIF.shift == "ARTERY") {
     dev.new(width = 6, height = 6, xpos = 1500,
     ypos = 0)
     plot(map.times, aif, ylim = c(0, max(aif,
     aif.shifted)), xlab = "min", ylab = "contrast agent",
     main = paste("Shifted Arterial Input Function (",
     format(param.est.whole.roi.xTofts$tlag *
     60, digits = 3), " sec)", sep = ""),
     type = "n")
     lines(map.times, aif, col = "red")
     lines(map.times, aif.shifted)
     legend(x = 0.55 * max(map.times), y = max(aif,
     aif.shifted), c("raw AIF", "shifted AIF"),
     c("red", "black"))
     }
     if (AIF.shift == "VEIN") {
     dev.new(width = 6, height = 6, xpos = 1500,
     ypos = 0)
     plot(map.times, aif, ylim = c(0, max(aif,
     aif.shifted)), xlab = "min", ylab = "contrast agent",
     main = paste("Shifted Venous Input Function (",
     format(param.est.whole.roi.xTofts$tlag *
     60, digits = 3), " sec)", sep = ""),
     type = "n")
     lines(map.times, aif, col = "blue")
     lines(map.times, aif.shifted)
     legend(x = 0.55 * max(map.times), y = max(aif,
     aif.shifted), c("raw VIF", "shifted VIF"),
     c("blue", "black"))
     }
     if (AIF.shift == "NONE") {
     dev.new(width = 6, height = 6, xpos = 1500,
     ypos = 0)
     plot(map.times, aif, ylim = c(0, max(aif)),
     xlab = "min", ylab = "contrast agent",
     main = "Vascular Input Function", type = "n")
     lines(map.times, aif)
     legend(x = 0.55 * max(map.times), y = max(aif),
     c("raw VIF"), c("black"))
     }
     }
     if (show.rt.fits == TRUE) {
     dev.new(width = 6, height = 6, xpos = 1500,
     ypos = 575)
     plot(map.times, roi, xlab = "min", ylab = "contrast agent",
     main = "median contrast agent conc; Tofts=blue, xTofts=red",
     cex = 2)
     text(x = 0.65 * max(map.times), y = 0.4 * max(roi,
     na.rm = TRUE), labels = "EXTENDED TOFTS PARAMS",
     cex = 1)
     if (class(fit.roi.median.xTofts) != "try-error") {
     if (fit.roi.median.xTofts$convergence ==
     0) {
     text(x = 0.65 * max(map.times), y = 0.35 *
     max(roi, na.rm = TRUE), labels = paste(expression(K^trans),
     " = ", format(param.est.whole.roi.xTofts$Ktrans,
     digits = 3), " 1/min (", cv.whole.roi.xTofts[1],
     "%)", sep = ""), cex = 1)
     text(x = 0.65 * max(map.times), y = 0.3 *
     max(roi, na.rm = TRUE), labels = paste(expression(k_ep),
     " = ", format(param.est.whole.roi.xTofts$kep,
     digits = 3), " 1/min (", cv.whole.roi.xTofts[2],
     "%)", sep = ""), cex = 1)
     text(x = 0.65 * max(map.times), y = 0.25 *
     max(roi, na.rm = TRUE), labels = paste(expression(v_b),
     " = ", format(param.est.whole.roi.xTofts$vb,
     digits = 3), " dimensionless (", cv.whole.roi.xTofts[3],
     "%)", sep = ""), cex = 1)
     if (AIF.shift != "NONE")
     text(x = 0.65 * max(map.times), y = 0.2 *
     max(roi, na.rm = TRUE), labels = paste(expression(t_lag),
     " = ", format(60 * param.est.whole.roi.xTofts$tlag,
     digits = 3), " sec (", cv.whole.roi.xTofts[4],
     "%)", sep = ""), cex = 1)
     lines(map.times, roi.median.fitted.xTofts,
     col = "red", lwd = 5)
     }
     }
     if (class(fit.roi.median.Tofts) != "try-error") {
     if (fit.roi.median.Tofts$convergence == 0) {
     lines(map.times, roi.median.fitted.Tofts,
     col = "blue", lwd = 2)
     }
     }
     }
     }
     if (file.original == FALSE) {
     fix.tlag <- TRUE
     if (file.format == "matlab") {
     mat_data <- readMat(file)
     args <- mat_data$args[, , 1]
     results_file <- args$resultsfile
     ID.visit <- args$IDvisit
     slice <- args$slice
     method.optimization <- args$methodoptimization
     show.rt.fits <- args$showrtfits
     param.for.avdt <- param.for.avdt
     range.map <- args$rangemap
     cutoff.map <- args$cutoffmap
     lo <- args$lo
     est.per.voxel.tlag <- args$estpervoxeltlag
     p0.xT <- mat_data$p0xT
     p0.T <- mat_data$p0T
     AIF.shift <- args$AIFshift
     nx <- mat_data$nx
     ny <- mat_data$ny
     nt <- mat_data$nt
     map_cc_slice <- mat_data$cc
     map_cc_roi <- mat_data$ccroi
     map.times <- mat_data$maptimes
     aif <- mat_data$aif
     aif.shifted <- mat_data$aifshifted
     map.KtransxT <- mat_data$mapKtransxT
     map.tlagxT <- mat_data$maptlagxT
     map.kepxT <- mat_data$mapkepxT
     map.vexT <- mat_data$mapvexT
     map.vbxT <- mat_data$mapvbxT
     map.KtransT.cv <- mat_data$mapKtransTcv
     map.kepT.cv <- mat_data$mapkepTcv
     map.KtransxT.cv <- mat_data$mapKtransxTcv
     map.tlagxT.cv <- mat_data$maptlagxTcv
     map.kepxT.cv <- mat_data$mapkepxTcv
     map.vbxT.cv <- mat_data$mapvbxTcv
     map.fitfailuresxT <- mat_data$mapfitfailuresxT
     map.KtransT <- mat_data$mapKtransT
     map.kepT <- mat_data$mapkepT
     map.veT <- mat_data$mapveT
     map.fitfailuresT <- mat_data$mapfitfailuresT
     mask.roi <- mat_data$maskroi
     param.est.medianT <- mat_data$paramestmedianT
     param.est.medianxT <- mat_data$paramestmedianxT
     cc.median <- mat_data$ccmedian
     roi.median.fitted <- mat_data$roimedianfitted
     param.est.whole.roi <- mat_data$paramestwholeroi
     map.AIC.compare <- mat_data$mapAICcompare
     map.AIC.T <- mat_data$mapAICT
     map.EF <- mat_data$mapEF
     map.AIC.xT <- mat_data$mapAICxT
     roi.median.fitted.Tofts <- mat_data$roimedianfittedTofts
     param.est.whole.roi.Tofts <- mat_data$paramestwholeroiTofts
     roi.median.fitted.xTofts <- mat_data$roimedianfittedxTofts
     param.est.whole.roi.xTofts <- mat_data$paramestwholeroixTofts
     cv.whole.roi.xTofts <- mat_data$cvwholeroixTofts
     params.xTofts <- mat_data$paramsxTofts
     rm(mat_data)
     }
     if (file.format == "RData") {
     load(file)
     args <- dcemri.data$args
     results_file <- args$resultsfile
     ID.visit <- args$IDvisit
     slice <- args$slice
     method.optimization <- args$methodoptimization
     show.rt.fits <- args$showrtfits
     param.for.avdt <- param.for.avdt
     range.map <- args$rangemap
     cutoff.map <- args$cutoffmap
     lo <- args$lo
     p0.xT <- dcemri.data$p0xT
     p0.T <- dcemri.data$p0T
     AIF.shift <- args$AIFshift
     est.per.voxel.tlag <- args$estpervoxeltlag
     nx <- dcemri.data$nx
     ny <- dcemri.data$ny
     nt <- dcemri.data$nt
     export.matlab <- args$exportmatlab
     map_cc_slice <- dcemri.data$cc
     map_cc_roi <- dcemri.data$ccroi
     map.times <- dcemri.data$maptimes
     aif <- dcemri.data$aif
     aif.shifted <- dcemri.data$aifshifted
     map.KtransT.cv <- dcemri.data$mapKtransTcv
     map.kepT.cv <- dcemri.data$mapkepTcv
     map.KtransxT.cv <- dcemri.data$mapKtransxTcv
     map.tlagxT.cv <- dcemri.data$maptlagxTcv
     map.kepxT.cv <- dcemri.data$mapkepxTcv
     map.vbxT.cv <- dcemri.data$mapvbxTcv
     map.KtransxT <- dcemri.data$mapKtransxT
     map.tlagxT <- dcemri.data$maptlagxT
     map.kepxT <- dcemri.data$mapkepxT
     map.vexT <- dcemri.data$mapvexT
     map.vbxT <- dcemri.data$mapvbxT
     map.fitfailuresxT <- dcemri.data$mapfitfailuresxT
     map.KtransT <- dcemri.data$mapKtransT
     map.kepT <- dcemri.data$mapkepT
     map.veT <- dcemri.data$mapveT
     map.fitfailuresT <- dcemri.data$mapfitfailuresT
     mask.roi <- dcemri.data$maskroi
     param.est.medianT <- dcemri.data$paramestmedianT
     param.est.medianxT <- dcemri.data$paramestmedianxT
     cc.median <- dcemri.data$ccmedian
     roi.median.fitted <- dcemri.data$roimedianfitted
     param.est.whole.roi <- dcemri.data$paramestwholeroi
     map.AIC.compare <- dcemri.data$mapAICcompare
     map.AIC.T <- dcemri.data$mapAICT
     map.EF <- dcemri.data$mapEF
     map.AIC.xT <- dcemri.data$mapAICxT
     roi.median.fitted.Tofts <- dcemri.data$roimedianfittedTofts
     param.est.whole.roi.Tofts <- dcemri.data$paramestwholeroiTofts
     roi.median.fitted.xTofts <- dcemri.data$roimedianfittedxTofts
     param.est.whole.roi.xTofts <- dcemri.data$paramestwholeroixTofts
     cv.whole.roi.xTofts <- dcemri.data$cvwholeroixTofts
     params.xTofts <- dcemri.data$paramsxTofts
     rm(dcemri.data)
     }
     modeltype1 <- "xTofts"
     modeltype2 <- "Tofts"
     }
     if (file.original == TRUE) {
     cat("fitting", modeltype1, "and", modeltype2,
     "models to ROI voxels...", "\n")
     ptm <- proc.time()[3]
     t <- map.times
     map.KtransxT <- matrix(NA, nrow = nx, ncol = ny)
     map.kepxT <- matrix(NA, nrow = nx, ncol = ny)
     map.vexT <- matrix(NA, nrow = nx, ncol = ny)
     map.vbxT <- matrix(NA, nrow = nx, ncol = ny)
     map.tlagxT <- matrix(NA, nrow = nx, ncol = ny)
     map.KtransxT.cv <- matrix(NA, nrow = nx, ncol = ny)
     map.kepxT.cv <- matrix(NA, nrow = nx, ncol = ny)
     map.vbxT.cv <- matrix(NA, nrow = nx, ncol = ny)
     map.tlagxT.cv <- matrix(NA, nrow = nx, ncol = ny)
     map.fitfailuresxT <- matrix(NA, nrow = nx, ncol = ny)
     map.OptimValuexT <- matrix(NA, nrow = nx, ncol = ny)
     map.KtransT <- matrix(NA, nrow = nx, ncol = ny)
     map.kepT <- matrix(NA, nrow = nx, ncol = ny)
     map.veT <- matrix(NA, nrow = nx, ncol = ny)
     map.KtransT.cv <- matrix(NA, nrow = nx, ncol = ny)
     map.kepT.cv <- matrix(NA, nrow = nx, ncol = ny)
     map.fitfailuresT <- matrix(NA, nrow = nx, ncol = ny)
     map.OptimValueT <- matrix(NA, nrow = nx, ncol = ny)
     map.AIC.compare <- matrix(0, nrow = nx, ncol = ny)
     map.AIC.T <- matrix(NA, nrow = nx, ncol = ny)
     map.EF <- matrix(NA, nrow = nx, ncol = ny)
     map.AIC.xT <- matrix(NA, nrow = nx, ncol = ny)
     cc_fittedxT <- array(0, dim = c(nx, ny, nt))
     cc_fittedT <- array(0, dim = c(nx, ny, nt))
     nv <- 1
     nv1_q <- trunc(quantile(1:length(mask.roi[mask.roi ==
     1]), probs = c(0.1, 0.2, 0.3, 0.4, 0.5, 0.6,
     0.7, 0.8, 0.9, 1)))
     if (show.rt.fits == TRUE)
     dev.new(xpos = 3500, ypos = 0)
     ptm_slice <- proc.time()[3]
     GTzero <- function(x) {
     length(as.vector(x > 0)[as.vector(x > 0) ==
     TRUE])/length(x)
     }
     for (x in 1:nx) {
     for (y in 1:ny) {
     if (mask.roi[x, y] == 1 & GTzero(map_cc_slice[x,
     y, ]) <= fracGTzero) {
     map.fitfailuresxT[x, y] <- -2
     map.fitfailuresT[x, y] <- -2
     }
     if (mask.roi[x, y] == 1 & GTzero(map_cc_slice[x,
     y, ]) > fracGTzero)
     map.EF[x, y] <- 1
     if (mask.roi[x, y] == 1 & GTzero(map_cc_slice[x,
     y, ]) > fracGTzero) {
     nv <- nv + 1
     roi <- map_cc_slice[x, y, ]
     fix.tlag <- TRUE
     roi.model <- roi.modelxT
     if (verbose == TRUE) {
     cat("x =", x, "\n")
     cat("y =", y, "\n")
     cat("contrast agent curve =", roi, "\n")
     cat("fitting xTofts model to voxel data...")
     }
     if (method.optimization == "L-BFGS-B")
     fit_roi <- try(optim(p0.xT, Obj_roi,
     model = roi.model, t = map.times, dt = diff(map.times),
     cp = aif, roi = roi, method = "L-BFGS-B",
     lower = c(lo, lo, lo), upper = c(Inf,
     Inf, Inf), hessian = TRUE), silent = try.silent)
     if (method.optimization != "L-BFGS-B")
     fit_roi <- try(optim(p0.xT, Obj_roi,
     model = roi.model, t = map.times, dt = diff(map.times),
     cp = aif, roi = roi, method = method.optimization,
     hessian = TRUE), silent = try.silent)
     if (verbose == TRUE)
     cat("done", "\n")
     if (class(fit_roi) != "try-error") {
     if (fit_roi$convergence == 0) {
     RSS <- sum((roi.model(p = fit_roi$par,
     t = map.times, dt = diff(map.times),
     cp = aif) - roi)^2)
     param_est <- fit_roi$par
     nD <- nt
     nP <- length(param_est)
     hess <- fit_roi$hessian
     cov <- try((nP * RSS/(nD - nP)) * solve(hess),
     silent = try.silent)
     if (class(cov) != "try-error") {
     sd <- try(sqrt(diag(cov)), silent = try.silent)
     if (class(sd) != "try-error") {
     cv <- sd/param_est * 100
     cv.roi.voxel <- as.numeric(format(cv,
     digits = 1))
     map.KtransxT.cv[x, y] <- cv.roi.voxel[1]
     map.kepxT.cv[x, y] <- cv.roi.voxel[2]
     map.vbxT.cv[x, y] <- cv.roi.voxel[3]
     if (est.per.voxel.tlag == TRUE)
     map.tlagxT.cv[x, y] <- cv.roi.voxel[4]
     }
     }
     map.KtransxT[x, y] <- fit_roi$par[1]
     map.kepxT[x, y] <- fit_roi$par[2]
     if (map.kepxT[x, y] < 1e-05)
     map.kepxT[x, y] <- 1e-05
     map.vexT[x, y] <- fit_roi$par[1]/map.kepxT[x,
     y]
     map.vbxT[x, y] <- fit_roi$par[3]
     if (est.per.voxel.tlag == TRUE)
     map.tlagxT[x, y] <- fit_roi$par[4]
     map.OptimValuexT[x, y] <- fit_roi$value
     }
     }
     if (class(fit_roi) == "try-error") {
     if (verbose == TRUE)
     cat("...but a <try-error> occurred",
     "\n")
     map.fitfailuresxT[x, y] <- 99
     }
     if (class(fit_roi) != "try-error")
     map.fitfailuresxT[x, y] <- fit_roi$convergence
     if (class(fit_roi) == "try-error") {
     map.KtransxT[x, y] <- NA
     map.kepxT[x, y] <- NA
     map.vexT[x, y] <- NA
     map.vbxT[x, y] <- NA
     map.tlagxT[x, y] <- NA
     }
     if (class(fit_roi) != "try-error") {
     if (fit_roi$convergence > 0) {
     map.KtransxT[x, y] <- NA
     map.kepxT[x, y] <- NA
     map.vexT[x, y] <- NA
     map.vbxT[x, y] <- NA
     map.tlagxT[x, y] <- NA
     }
     }
     if (verbose == TRUE)
     cat("simulating xTofts model at estimated parameter values...")
     if (class(fit_roi) != "try-error") {
     if (fit_roi$convergence == 0) {
     simulation <- roi.model(p = fit_roi$par,
     t = map.times, dt = diff(map.times),
     cp = aif)
     cc_fittedxT[x, y, ] <- simulation
     }
     }
     if (verbose == TRUE)
     cat("done", "\n")
     roi.model <- roi.modelT
     if (verbose == TRUE)
     cat("fitting Tofts model to voxel data...")
     if (method.optimization == "L-BFGS-B")
     fit_roiT <- try(optim(p0.T, Obj_roi,
     model = roi.model, t = map.times, dt = diff(map.times),
     cp = aif, roi = roi, method = "L-BFGS-B",
     lower = lower.wholeT, upper = upper.wholeT,
     hessian = TRUE), silent = try.silent)
     if (method.optimization != "L-BFGS-B")
     fit_roiT <- try(optim(p0.T, Obj_roi,
     model = roi.model, t = map.times, dt = diff(map.times),
     cp = aif, roi = roi, method = method.optimization,
     hessian = TRUE), silent = try.silent)
     if (verbose == TRUE)
     cat("done", "\n")
     if (class(fit_roiT) != "try-error") {
     if (fit_roiT$convergence == 0) {
     RSS <- sum((roi.model(p = fit_roiT$par,
     t = map.times, dt = diff(map.times),
     cp = aif) - roi)^2)
     nD <- nt
     nP <- length(param_est)
     param_est <- fit_roiT$par
     df <- nt - length(param_est)
     hess <- fit_roiT$hessian
     cov <- try((nP * RSS/(nD - nP)) * solve(hess),
     silent = try.silent)
     if (class(cov) != "try-error") {
     sd <- try(sqrt(diag(cov)), silent = try.silent)
     if (class(sd) != "try-error") {
     cv <- sd/param_est * 100
     cv.roi.voxel <- as.numeric(format(cv,
     digits = 1))
     map.KtransT.cv[x, y] <- cv.roi.voxel[1]
     map.kepT.cv[x, y] <- cv.roi.voxel[2]
     }
     }
     map.KtransT[x, y] <- fit_roiT$par[1]
     map.kepT[x, y] <- fit_roiT$par[2]
     if (map.kepT[x, y] < 1e-05)
     map.kepT[x, y] <- 1e-05
     map.veT[x, y] <- fit_roiT$par[1]/map.kepT[x,
     y]
     map.OptimValueT[x, y] <- fit_roiT$value
     }
     }
     if (class(fit_roiT) == "try-error") {
     map.fitfailuresT[x, y] <- 99
     if (verbose == TRUE)
     cat("...but a <try-error> occurred",
     "\n")
     }
     if (class(fit_roiT) == "try-error") {
     map.KtransT[x, y] <- NA
     map.kepT[x, y] <- NA
     map.veT[x, y] <- NA
     }
     if (class(fit_roiT) != "try-error") {
     if (fit_roiT$convergence > 0) {
     map.KtransT[x, y] <- NA
     map.kepT[x, y] <- NA
     map.veT[x, y] <- NA
     }
     }
     if (class(fit_roiT) != "try-error")
     map.fitfailuresT[x, y] <- fit_roiT$convergence
     if (verbose == TRUE)
     cat("simulating Tofts model at estimated parameter values...")
     if (class(fit_roiT) != "try-error") {
     if (fit_roiT$convergence == 0) {
     simulation_2 <- roi.model(p = fit_roiT$par,
     t = map.times, dt = diff(map.times),
     cp = aif)
     cc_fittedT[x, y, ] <- simulation_2
     }
     }
     if (verbose == TRUE)
     cat("done", "\n")
     roi.model <- roi.modelxT
     if (class(fit_roi) != "try-error" & class(fit_roiT) !=
     "try-error") {
     if (fit_roi$convergence == 0 & fit_roiT$convergence ==
     0) {
     if (verbose == TRUE)
     cat("calculating AICc values...")
     simulation_AIC <- simulation
     simulation_AIC_2 <- simulation_2
     np_1 <- length(fit_roiT$par)
     np_2 <- length(fit_roi$par) + 1
     AIC1 <- nt * log(fit_roiT$value) +
     2 * (np_1 + 1) + (2 * (np_1 + 1) *
     (np_1 + 2))/(nt - np_1 - 2)
     AIC2 <- nt * log(fit_roi$value) + 2 *
     (np_2 + 1) + (2 * (np_2 + 1) * (np_2 +
     2))/(nt - np_2 - 2)
     AIC1 <- as.numeric(format(AIC1, digits = 1))
     AIC2 <- as.numeric(format(AIC2, digits = 1))
     if (AIC2 < AIC1)
     map.AIC.compare[x, y] <- 1
     if (AIC2 >= AIC1)
     map.AIC.compare[x, y] <- 2
     map.AIC.T[x, y] <- AIC1
     map.AIC.xT[x, y] <- AIC2
     }
     }
     if (class(fit_roi) != "try-error") {
     if (fit_roi$convergence > 0) {
     map.AIC.compare[x, y] <- NA
     }
     }
     if (class(fit_roiT) != "try-error") {
     if (fit_roiT$convergence > 0) {
     map.AIC.compare[x, y] <- NA
     }
     }
     if (class(fit_roi) == "try-error" | class(fit_roiT) ==
     "try-error")
     map.AIC.compare[x, y] <- NA
     if (verbose == TRUE) {
     cat("done", "\n")
     cat("======================", "\n")
     }
     if (show.rt.fits == TRUE) {
     if (class(fit_roiT) != "try-error" &
     class(fit_roi) != "try-error") {
     if (fit_roiT$convergence == 0 & fit_roi$convergence ==
     0) {
     plot(map.times, roi, ylab = "contrast agent",
     xlab = "min", main = paste(modeltype1,
     "(red) and", modeltype2, "(blue)"),
     cex = 3)
     lines(map.times, simulation, col = "red",
     lwd = 5)
     lines(map.times, simulation_2, col = "blue",
     lwd = 2)
     }
     }
     }
     if (nv == 2)
     cat("progress: ")
     if (nv == nv1_q[[1]])
     cat("10%..")
     if (nv == nv1_q[[2]])
     cat("20%..")
     if (nv == nv1_q[[3]])
     cat("30%..")
     if (nv == nv1_q[[4]])
     cat("40%..")
     if (nv == nv1_q[[5]])
     cat("50%..")
     if (nv == nv1_q[[6]])
     cat("60%..")
     if (nv == nv1_q[[7]])
     cat("70%..")
     if (nv == nv1_q[[8]])
     cat("80%..")
     if (nv == nv1_q[[9]])
     cat("90%..", "\n")
     if (nv == nv1_q[[10]] - 10)
     cat("..10")
     if (nv == nv1_q[[10]] - 9)
     cat("..9..")
     if (nv == nv1_q[[10]] - 8)
     cat("8..")
     if (nv == nv1_q[[10]] - 7)
     cat("7..")
     if (nv == nv1_q[[10]] - 6)
     cat("6..")
     if (nv == nv1_q[[10]] - 5)
     cat("5..")
     if (nv == nv1_q[[10]] - 4)
     cat("4..")
     if (nv == nv1_q[[10]] - 3)
     cat("3..")
     if (nv == nv1_q[[10]] - 2)
     cat("2..")
     if (nv == nv1_q[[10]] - 1)
     cat("1..", "\n")
     }
     }
     }
     cat("..done in", format((proc.time()[3] - ptm)/60,
     digits = 2), "minutes.", "\n")
     cat("--------", "\n")
     graphics.off()
     KtransxT.median <- median(map.KtransxT[map.KtransxT >
     0 & map.fitfailuresxT == 0], na.rm = TRUE)
     kepxT.median <- median(map.kepxT[map.kepxT >
     0 & map.fitfailuresxT == 0], na.rm = TRUE)
     vexT.median <- median(map.vexT[map.vexT > 0 &
     map.fitfailuresxT == 0], na.rm = TRUE)
     vbxT.median <- median(map.vbxT[map.vbxT >= 0 &
     map.fitfailuresxT == 0], na.rm = TRUE)
     if (est.per.voxel.tlag == TRUE)
     tlagxT.median <- median(map.tlagxT[map.tlagxT >=
     0 & map.fitfailuresxT == 0], na.rm = TRUE)
     if (est.per.voxel.tlag == FALSE)
     tlagxT.median <- NA
     fitfailuresxT.total <- length(map.fitfailuresxT[map.fitfailuresxT >=
     1])/length(map.fitfailuresxT[map.fitfailuresxT >=
     0]) * 100
     param.est.medianxT <- list(KtransxT.median, kepxT.median,
     vexT.median, vbxT.median, tlagxT.median, fitfailuresxT.total)
     names(param.est.medianxT) <- c("Ktrans.median",
     "kep.median", "ve.median", "vb.median", "tlag.median",
     "percent.fitfailures")
     KtransT.median <- median(map.KtransT[map.KtransT >
     0 & map.fitfailuresT == 0], na.rm = TRUE)
     kepT.median <- median(map.kepT[map.kepT > 0 &
     map.fitfailuresT == 0], na.rm = TRUE)
     veT.median <- median(map.veT[map.veT > 0 & map.fitfailuresT ==
     0], na.rm = TRUE)
     fitfailuresT.total <- length(map.fitfailuresT[map.fitfailuresT >=
     1])/length(map.fitfailuresT[map.fitfailuresT >=
     0]) * 100
     param.est.medianT <- list(KtransT.median, kepT.median,
     veT.median, fitfailuresT.total)
     names(param.est.medianT) <- c("Ktrans.median",
     "kep.median", "ve.median", "percent.fitfailures")
     }
     if (file.original == TRUE) {
     if (file == "concatenate.KAT.with.KAT.checkData.RData")
     ID.visit <- paste(strsplit(results_file, split = ".mat")[[1]],
     "_s", slice, sep = "")
     if (file != "concatenate.KAT.with.KAT.checkData.RData")
     ID.visit <- paste(strsplit(file, split = ".mat")[[1]],
     "_s", slice, sep = "")
     }
     if (file.original == FALSE)
     ID.visit <- strsplit(file, split = ".mat")[[1]]
     ID.visit <- strsplit(ID.visit, split = "/")
     ID.visit <- ID.visit[[1]][length(ID.visit[[1]])]
     IDvp <- strsplit(ID.visit, split = "_")
     ID.visit.forplot <- paste(IDvp[[1]][1], ".", IDvp[[1]][2],
     ".", IDvp[[1]][3], ".", IDvp[[1]][4], sep = "")
     DATE <- date()
     if (length(strsplit(DATE, split = " ")[[1]]) ==
     2) {
     full_date <- strsplit(DATE, split = " ")[[1]]
     full_date_1 <- full_date[1]
     full_date_1 <- strsplit(full_date_1, split = " ")[[1]]
     full_date_2 <- full_date[2]
     full_date_2 <- strsplit(full_date_2, split = " ")[[1]]
     month <- full_date_1[2]
     day <- full_date_2[1]
     time <- full_date_2[2]
     year <- full_date_2[3]
     }
     if (length(strsplit(DATE, split = " ")[[1]]) ==
     1) {
     full_date <- strsplit(DATE, split = " ")[[1]]
     month <- full_date[2]
     day <- full_date[3]
     time <- full_date[4]
     year <- full_date[5]
     }
     year <- strsplit(year, split = "")[[1]]
     year <- paste(year[3], year[4], sep = "")
     time_concat <- strsplit(time, split = ":")[[1]]
     time_concat <- paste(paste(time_concat[1], time_concat[2],
     sep = ""), time_concat[3], sep = "")
     DATE <- paste(paste(paste(paste(day, month, sep = ""),
     year, sep = ""), "-", sep = ""), time_concat,
     sep = "")
     filename3 <- paste(ID.visit, "_KAT_", DATE, ".mat",
     sep = "")
     filename3 <- sub(".RData", "", filename3)
     if (file.original == FALSE) {
     roi.model <- roi.modelxT
     }
     if (param.for.avdt == "Ktrans")
     MAP <- map.KtransxT
     if (param.for.avdt == "kep")
     MAP <- map.kepxT
     if (param.for.avdt == "ve")
     MAP <- map.vexT
     if (param.for.avdt == "vb")
     MAP <- map.vbxT
     if (param.for.avdt == "tlag")
     MAP <- map.tlagxT
     x_min <- 0
     x_max <- nx
     y_min <- 0
     y_max <- ny
     for (xx in 1:nx) {
     if (sum(MAP[xx, ], na.rm = TRUE) > 0) {
     x_min <- xx - 3
     break
     }
     if (sum(MAP[xx, ], na.rm = TRUE) == 0)
     xx <- xx + 1
     }
     for (y in 1:ny) {
     if (sum(MAP[, y], na.rm = TRUE) > 0) {
     y_min <- y - 3
     break
     }
     if (sum(MAP[, y], na.rm = TRUE) == 0)
     y <- y + 1
     }
     for (xx in nx:0) {
     if (sum(MAP[xx, ], na.rm = TRUE) > 0) {
     x_max <- xx + 3
     break
     }
     if (sum(MAP[xx, ], na.rm = TRUE) == 0)
     xx <- xx - 1
     }
     for (y in ny:0) {
     if (sum(MAP[, y], na.rm = TRUE) > 0) {
     y_max <- y + 3
     break
     }
     if (sum(MAP[, y], na.rm = TRUE) == 0)
     y <- y - 1
     }
     MAP_ul_s1 <- MAP[MAP > 0]
     MAP_ul_s1 <- sort(MAP_ul_s1)
     MAP_ul <- range.map * (max(MAP_ul_s1[1:length(MAP_ul_s1) *
     cutoff.map]))
     if (file.original == FALSE) {
     if (param.for.avdt == "Ktrans")
     MAP <- map.KtransxT
     if (param.for.avdt == "kep")
     MAP <- map.kepxT
     if (param.for.avdt == "ve")
     MAP <- map.vexT
     if (param.for.avdt == "vb")
     MAP <- map.vbxT
     if (param.for.avdt == "tlag")
     MAP <- map.tlagxT
     if (file.format == "matlab") {
     if (param.for.avdt == "Ktrans")
     param.median <- param.est.medianxT[, , 1]$Ktrans.median[1]
     if (param.for.avdt == "kep")
     param.median <- param.est.medianxT[, , 1]$kep.median[1]
     if (param.for.avdt == "ve")
     param.median <- param.est.medianxT[, , 1]$ve.median[1]
     if (param.for.avdt == "vb")
     param.median <- param.est.medianxT[, , 1]$vb.median[1]
     if (param.for.avdt == "tlag")
     param.median <- param.est.medianxT[, , 1]$tlag.median[1]
     }
     if (file.format == "RData") {
     if (param.for.avdt == "Ktrans")
     param.median <- param.est.medianxT$Ktrans.median
     if (param.for.avdt == "kep")
     param.median <- param.est.medianxT$kep.median
     if (param.for.avdt == "ve")
     param.median <- param.est.medianxT$ve.median
     if (param.for.avdt == "vb")
     param.median <- param.est.medianxT$vb.median
     if (param.for.avdt == "tlag")
     param.median <- param.est.medianxT$tlag.median
     }
     MAP_for_plot <- MAP
     MAP_for_plot[MAP_for_plot < 0] <- 0
     MAP_for_plot[MAP_for_plot >= MAP_ul * 0.99] <- MAP_ul *
     0.99
     dev.new(width = 6, height = 6, xpos = 1860, ypos = 578)
     plot(map.times, cc.median, xlab = "min", ylab = "contrast agent",
     main = paste(modeltype1, "(red) and", modeltype2,
     "(blue) fitted to median whole ROI data"),
     cex.main = 1, cex.axis = 1, cex.lab = 1, cex = 2)
     lines(map.times, roi.median.fitted.xTofts, col = "red",
     lwd = 5)
     lines(map.times, roi.median.fitted.Tofts, col = "blue",
     lwd = 2)
     text(x = 0.6 * max(map.times), y = 0.3 * max(cc.median,
     na.rm = TRUE), labels = "EXTENDED TOFTS PARAMS",
     cex = 1)
     text(x = 0.6 * max(map.times), y = 0.24 * max(cc.median,
     na.rm = TRUE), labels = paste(expression(K^trans),
     " = ", format(param.est.whole.roi.xTofts$Ktrans,
     digits = 3), " 1/min (", cv.whole.roi.xTofts[1],
     "%)", sep = ""), cex = 1)
     text(x = 0.6 * max(map.times), y = 0.18 * max(cc.median,
     na.rm = TRUE), labels = paste(expression(k_ep),
     " = ", format(param.est.whole.roi.xTofts$kep,
     digits = 3), " 1/min (", cv.whole.roi.xTofts[2],
     "%)", sep = ""), cex = 1)
     text(x = 0.6 * max(map.times), y = 0.12 * max(cc.median,
     na.rm = TRUE), labels = paste(expression(v_b),
     " = ", format(param.est.whole.roi.xTofts$vb,
     digits = 3), " dimensionless (", cv.whole.roi.xTofts[3],
     "%)", sep = ""), cex = 1)
     if (AIF.shift != "NONE")
     text(x = 0.6 * max(map.times), y = 0.06 * max(cc.median,
     na.rm = TRUE), labels = paste(expression(t_lag),
     " = ", format(60 * param.est.whole.roi.xTofts$tlag,
     digits = 3), " sec (", cv.whole.roi.xTofts[4],
     "%)", sep = ""), cex = 1)
     dev.new(width = 6, height = 6, xpos = 1860, ypos = 0)
     image(map.AIC.compare, xlim = c(x_min/nx, x_max/nx),
     ylim = c(y_min/ny, y_max/ny), col = palette(colorRampPalette(c("black",
     "red", "blue"))(3)), main = paste("best fit:",
     modeltype1, "(red) and", modeltype2, "(blue)"))
     image(map.AIC.compare, xlim = c(x_min/nx, x_max/nx),
     ylim = c(y_min/ny, y_max/ny), col = palette(colorRampPalette(c("black",
     "red", "blue"))(3)), main = paste("best fit:",
     modeltype1, "(red) and", modeltype2, "(blue)"))
     dev.new(width = 12.75, height = 12.75, xpos = 238,
     ypos = 0)
     image(MAP_for_plot, col = palette(colorRampPalette(c("black",
     "red", "yellow"))(1000)), xlim = c(x_min/nx,
     x_max/nx), ylim = c(y_min/ny, y_max/ny), zlim = c(0,
     MAP_ul), main = paste("Model Type=", modeltype1,
     " Median ", param.for.avdt, "=", format(param.median,
     digit = 2), " VisitID=", strsplit(args$IDvisit,
     split = "_")[[1]][1], " slice=", args$slice,
     sep = ""))
     image(MAP_for_plot, col = palette(colorRampPalette(c("black",
     "red", "yellow"))(1000)), xlim = c(x_min/nx,
     x_max/nx), ylim = c(y_min/ny, y_max/ny), zlim = c(0,
     MAP_ul), main = paste("Model Type=", modeltype1,
     " Median ", param.for.avdt, "=", format(param.median,
     digit = 2), " VisitID=", strsplit(args$IDvisit,
     split = "_")[[1]][1], " slice=", args$slice,
     sep = ""))
     text(x_min/nx + (x_max/nx - x_min/nx) * 0.97,
     y_min/ny + (y_max/ny - y_min/ny) * 0.99, "close",
     col = "green")
     text(x_min/nx + (x_max/nx - x_min/nx) * 0.05,
     y_min/ny + (y_max/ny - y_min/ny) * 0.99, "print to PDF",
     col = "green")
     text(x_min/nx + (x_max/nx - x_min/nx) * 0.84,
     y_min/ny + (y_max/ny - y_min/ny) * 0.01, paste("KAT for DCEMRI v",
     KAT.version, ", Genentech PTPK", sep = ""),
     col = "darkgrey")
     legend <- seq(0, MAP_ul, by = 0.001)
     dim(legend) <- c(1, length(legend))
     dev.new(width = 2.5, height = 12.75, xpos = 0,
     ypos = 0)
     image(legend, col = palette(colorRampPalette(c("black",
     "red", "yellow"))(1000)), zlim = c(0, MAP_ul),
     xaxt = "n", yaxt = "n", xlab = "0", main = format(MAP_ul,
     digits = 3), ylab = expression(min^-1), cex.lab = 1.25,
     cex.main = 1.05)
     image(legend, col = palette(colorRampPalette(c("black",
     "red", "yellow"))(1000)), zlim = c(0, MAP_ul),
     xaxt = "n", yaxt = "n", xlab = "0", main = format(MAP_ul,
     digits = 3), ylab = expression(min^-1), cex.lab = 1.25,
     cex.main = 1.05)
     if (AIF.shift == "ARTERY") {
     dev.new(width = 5.15, height = 4.4, xpos = 1500,
     ypos = 0)
     plot(map.times, aif, ylim = c(0, max(aif, aif.shifted)),
     xlab = "min", ylab = "contrast agent", main = paste("Shifted Arterial Input Function (",
     format(param.est.whole.roi.xTofts$tlag *
     60, digits = 3), " sec)", sep = ""),
     type = "n")
     lines(map.times, aif, col = "red")
     lines(map.times, aif.shifted)
     legend(x = 0.55 * max(map.times), y = max(aif,
     aif.shifted), c("raw AIF", "shifted AIF"),
     c("red", "black"))
     }
     if (AIF.shift == "VEIN") {
     dev.new(width = 5.15, height = 4.4, xpos = 1500,
     ypos = 0)
     plot(map.times, aif, ylim = c(0, max(aif, aif.shifted)),
     xlab = "min", ylab = "contrast agent", main = paste("Shifted Venous Input Function (",
     format(param.est.whole.roi.xTofts$tlag *
     60, digits = 3), " sec)", sep = ""),
     type = "n")
     lines(map.times, aif, col = "blue")
     lines(map.times, aif.shifted)
     legend(x = 0.55 * max(map.times), y = max(aif,
     aif.shifted), c("raw VIF", "shifted VIF"),
     c("blue", "black"))
     }
     if (AIF.shift == "NONE") {
     dev.new(width = 5.15, height = 4.4, xpos = 1500,
     ypos = 0)
     plot(map.times, aif, ylim = c(0, max(aif)),
     xlab = "min", ylab = "contrast agent", main = "Vascular Input Function",
     type = "n")
     lines(map.times, aif)
     legend(x = 0.55 * max(map.times), y = max(aif),
     c("raw VIF"), c("black"))
     }
     dev.set(4)
     inf <- 1
     newplot <- 1
     newplot2 <- 1
     legend_count <- 1
     legend_labels <- 1:1000
     legend_matrix <- matrix(0, ncol = ny, nrow = nx)
     cat("---", "\n")
     while (inf == 1) {
     z <- locator(1, type = "o", col = "green")
     xx <- round(z$x * (nx - 1) + 1)
     yy <- round(z$y * (ny - 1) + 1)
     if (legend_matrix[xx, yy] == 0) {
     legend(z$x, z$y, legend_labels[legend_count],
     col = "green", text.col = "green")
     legend_matrix[xx, yy] <- 1
     cat("Voxel Number/Coordinates: n=", legend_count,
     ", x=", xx, ", y=", yy, "\n", sep = "")
     cat("Parameter estimates (xTofts): Ktrans=",
     format(map.KtransxT[xx, yy], digits = 3),
     ", ve=", format(map.KtransxT[xx, yy]/map.kepxT[xx,
     yy], digits = 3), ", vb=", format(map.vbxT[xx,
     yy], digits = 3), ", tlag=", format(map.tlagxT[xx,
     yy], digits = 3), "\n", sep = "")
     cat("%CVs of Parameter estimates (xTofts): Ktrans=",
     format(map.KtransxT.cv[xx, yy], digits = 2),
     ", kep (Ktrans/ve)=", format(map.kepxT.cv[xx,
     yy], digits = 2), ", vb=", format(map.vbxT.cv[xx,
     yy], digits = 2), ", tlag=", format(map.tlagxT.cv[xx,
     yy], digits = 2), "\n", sep = "")
     cat("Parameter estimates (Tofts): Ktrans=",
     format(map.KtransT[xx, yy], digits = 3),
     ", ve=", format(map.KtransT[xx, yy]/map.kepT[xx,
     yy], digits = 3), "\n", sep = "")
     cat("%CVs of Parameter estimates (Tofts): Ktrans=",
     format(map.KtransT.cv[xx, yy], digits = 2),
     ", kep (Ktrans/ve)=", format(map.kepT.cv[xx,
     yy], digits = 3), "\n", sep = "")
     cat("---", "\n")
     legend_count <- legend_count + 1
     }
     xdim <- x_max - x_min
     ydim <- y_max - y_min
     if (xx > (x_max - 0.1 * xdim) && yy > (y_max -
     0.02 * ydim)) {
     graphics.off()
     dev.off()
     }
     xx_old <- xx
     yy_old <- yy
     conc <- 1:nt
     for (i in 1:nt) conc[i] <- map_cc_slice[xx,
     yy, i]
     if (xx < (x_min + 0.12 * xdim) && yy > (y_max -
     0.02 * ydim)) {
     pdf(file = paste(results_file, "-SUMMARY.pdf",
     sep = ""), height = 12, width = 15)
     layout(matrix(c(1, 2, 3, 4, 5, 6), 2, 3,
     byrow = TRUE), widths = c(1.5, 5.5, 5.5))
     par(omi = c(0.15, 0.15, 0.15, 0.15))
     image(legend, col = palette(colorRampPalette(c("black",
     "red", "yellow"))(1000)), zlim = c(0, MAP_ul),
     xaxt = "n", yaxt = "n", xlab = "0", main = format(MAP_ul,
     digits = 3), ylab = expression(min^-1),
     cex.lab = 1.5, cex.main = 1.5, cex.axis = 1.5)
     image(legend, col = palette(colorRampPalette(c("black",
     "red", "yellow"))(1000)), zlim = c(0, MAP_ul),
     xaxt = "n", yaxt = "n", xlab = "0", main = format(MAP_ul,
     digits = 3), ylab = expression(min^-1),
     cex.lab = 1.5, cex.main = 1.5, cex.axis = 1.5,
     add = TRUE)
     image(MAP_for_plot, col = palette(colorRampPalette(c("black",
     "red", "yellow"))(1000)), xlim = c(x_min/nx,
     x_max/nx), ylim = c(y_min/ny, y_max/ny),
     zlim = c(0, MAP_ul), main = paste("Model Type=",
     modeltype1, " Median ", param.for.avdt,
     "=", format(param.median, digit = 2),
     " VisitID=", strsplit(args$IDvisit,
     split = "_")[[1]][1], " slice=",
     args$slice, sep = ""), cex.axis = 1.5)
     image(MAP_for_plot, col = palette(colorRampPalette(c("black",
     "red", "yellow"))(1000)), xlim = c(x_min/nx,
     x_max/nx), ylim = c(y_min/ny, y_max/ny),
     zlim = c(0, MAP_ul), main = paste("Model Type=",
     modeltype1, " Median ", param.for.avdt,
     "=", format(param.median, digit = 2),
     " VisitID=", strsplit(args$IDvisit,
     split = "_")[[1]][1], " slice=",
     args$slice, sep = ""), cex.axis = 1.5,
     add = TRUE)
     plot(map.times, cc.median, xlab = "min",
     ylab = "contrast agent", main = "median contrast agent conc; Tofts=blue, xTofts=red",
     cex = 3, cex.axis = 1.5, cex.lab = 1.5)
     text(x = 0.6 * max(map.times), y = 0.25 *
     max(cc.median, na.rm = TRUE), labels = "EXTENDED TOFTS PARAMS",
     cex = 1.5)
     text(x = 0.6 * max(map.times), y = 0.2 *
     max(cc.median, na.rm = TRUE), labels = paste(expression(K^trans),
     " = ", format(param.est.whole.roi.xTofts$Ktrans,
     digits = 3), " 1/min (", cv.whole.roi.xTofts[1],
     "%)", sep = ""), cex = 1.5)
     text(x = 0.6 * max(map.times), y = 0.15 *
     max(cc.median, na.rm = TRUE), labels = paste(expression(v_e),
     " = ", format(param.est.whole.roi.xTofts$ve,
     digits = 3), " 1/min (", cv.whole.roi.xTofts[2],
     "%)", sep = ""), cex = 1.5)
     text(x = 0.6 * max(map.times), y = 0.1 *
     max(cc.median, na.rm = TRUE), labels = paste(expression(v_b),
     " = ", format(param.est.whole.roi.xTofts$vb,
     digits = 3), " dimensionless (", cv.whole.roi.xTofts[3],
     "%)", sep = ""), cex = 1.5)
     if (AIF.shift != "NONE")
     text(x = 0.6 * max(map.times), y = 0.05 *
     max(cc.median, na.rm = TRUE), labels = paste(expression(t_lag),
     " = ", format(60 * param.est.whole.roi.xTofts$tlag,
     digits = 3), " sec (", cv.whole.roi.xTofts[4],
     "%)", sep = ""), cex = 1.5)
     lines(map.times, roi.median.fitted.xTofts,
     col = "red", lwd = 5)
     lines(map.times, roi.median.fitted.Tofts,
     col = "blue", lwd = 2)
     text(x = 0.5 * max(map.times), 1 * max(cc.median,
     na.rm = TRUE), paste("R package version =",
     KAT.version), cex = 1.5)
     image(array(1:2, dim = c(1, 2)), col = palette(colorRampPalette(c("red",
     "blue"))(2)), zlim = c(0, 2), xaxt = "n",
     yaxt = "n", xlab = modeltype1, main = modeltype2,
     cex.lab = 1.5, cex.main = 1.5, cex.axis = 1.5)
     image(array(1:2, dim = c(1, 2)), col = palette(colorRampPalette(c("red",
     "blue"))(2)), zlim = c(0, 2), xaxt = "n",
     yaxt = "n", xlab = modeltype1, main = modeltype2,
     cex.lab = 1.5, cex.main = 1.5, cex.axis = 1.5,
     add = TRUE)
     image(map.AIC.compare, xlim = c(x_min/nx,
     x_max/nx), ylim = c(y_min/ny, y_max/ny),
     col = palette(colorRampPalette(c("black",
     "red", "blue"))(3)), main = "Summary of model discrimination analysis",
     cex.axis = 1.5, cex.lab = 1.5)
     image(map.AIC.compare, xlim = c(x_min/nx,
     x_max/nx), ylim = c(y_min/ny, y_max/ny),
     col = palette(colorRampPalette(c("black",
     "red", "blue"))(3)), main = "Summary of model discrimination analysis",
     cex.axis = 1.5, cex.lab = 1.5, add = TRUE)
     if (AIF.shift == "ARTERY") {
     plot(map.times, aif, ylim = c(0, max(aif,
     aif.shifted)), xlab = "min", ylab = "contrast agent",
     main = paste("Shifted Arterial Input Function (",
     format(param.est.whole.roi.xTofts$tlag *
     60, digits = 3), " sec)", sep = ""),
     type = "n", cex = 3, cex.axis = 1.5,
     cex.lab = 1.5)
     lines(map.times, aif, col = "red")
     lines(map.times, aif.shifted)
     legend(x = 0.55 * max(map.times), y = max(aif,
     aif.shifted), c("raw AIF", "shifted AIF"),
     c("red", "black"), cex = 2)
     }
     if (AIF.shift == "VEIN") {
     plot(map.times, aif, ylim = c(0, max(aif,
     aif.shifted)), xlab = "min", ylab = "contrast agent",
     main = paste("Shifted Venous Input Function (",
     format(param.est.whole.roi.xTofts$tlag *
     60, digits = 3), " sec)", sep = ""),
     type = "n", cex = 3, cex.axis = 1.5,
     cex.lab = 1.5)
     lines(map.times, aif, col = "blue")
     lines(map.times, aif.shifted)
     legend(x = 0.55 * max(map.times), y = max(aif,
     aif.shifted), c("raw VIF", "shifted VIF"),
     c("blue", "black"), cex = 2)
     }
     if (AIF.shift == "NONE") {
     plot(map.times, aif, ylim = c(0, max(aif)),
     xlab = "min", ylab = "contrast agent",
     main = "Vascular Input Function", type = "n",
     cex = 3, cex.axis = 1.5, cex.lab = 1.5)
     lines(map.times, aif)
     legend(x = 0.55 * max(map.times), y = max(aif),
     c("raw VIF"), c("black"), cex = 2)
     }
     dev.off()
     cat("image printed to pdf.", "\n")
     cat("---", "\n")
     }
     if (newplot == 1) {
     dev.new(width = 5.15, height = 4.4, xpos = 1500,
     ypos = 432)
     newplot <- 2
     }
     dev.set(7)
     plot(map.times, conc, xlab = "min", ylab = "contrast agent",
     ylim = c(-max(conc, na.rm = TRUE)/5, 1.4 *
     max(conc, na.rm = TRUE)), cex = 1.5, main = paste(paste("red=",
     modeltype1, ", blue=", modeltype2, " (",
     sep = ""), paste("x=", round(xx_old), ", y=",
     round(yy_old), ")", sep = ""), sep = ""))
     value_xTofts <- MAP[xx, yy]
     if (is.finite(value_xTofts) == FALSE)
     value_xTofts <- 0
     if (value_xTofts != 0) {
     if (est.per.voxel.tlag == TRUE)
     paramsxT <- c(map.KtransxT[xx, yy], map.kepxT[xx,
     yy], map.vbxT[xx, yy], map.tlagxT[xx,
     yy])
     if (est.per.voxel.tlag == FALSE)
     paramsxT <- c(map.KtransxT[xx, yy], map.kepxT[xx,
     yy], map.vbxT[xx, yy], param.est.whole.roi.xTofts$tlag)
     paramsT <- c(map.KtransT[xx, yy], map.kepT[xx,
     yy])
     roi.model <- roi.modelxT
     simulation <- roi.model(p = paramsxT, t = map.times,
     dt = diff(map.times), cp = aif)
     roi.model <- roi.modelT
     simulation_2 <- roi.model(p = paramsT, t = map.times,
     dt = diff(map.times), cp = aif)
     lines(map.times, simulation_2, col = "blue",
     lwd = 2)
     lines(map.times, simulation, col = "red",
     lwd = 2, lty = 2)
     roi.model <- roi.modelxT
     }
     text(max(map.times)/4.5, 1.4 * max(conc, na.rm = TRUE),
     "Fitted xTofts params")
     text(max(map.times)/4.5, 1.4 * max(conc, na.rm = TRUE) *
     0.93, paste(paste("Ktrans =", format(map.KtransxT[xx,
     yy], digits = 3)), "min^-1"))
     text(max(map.times)/4.5, 1.4 * max(conc, na.rm = TRUE) *
     0.86, paste("ve =", format(map.vexT[xx, yy],
     digits = 3)))
     text(max(map.times)/4.5, 1.4 * max(conc, na.rm = TRUE) *
     0.79, paste("vb =", format(map.vbxT[xx, yy],
     digits = 3)))
     if (est.per.voxel.tlag == TRUE)
     text(max(map.times)/4.5, 1.4 * max(conc,
     na.rm = TRUE) * 0.72, paste("tlag =", format(map.tlagxT[xx,
     yy], digits = 3)))
     if (newplot2 == 1) {
     dev.new(width = 5.15, height = 3, xpos = 1500,
     ypos = 864)
     newplot2 <- 2
     }
     else dev.set(8)
     if (value_xTofts != 0) {
     simulation_AIC <- simulation
     simulation_AIC_2 <- simulation_2
     if (length(simulation) == length(conc)) {
     plot(map.times, conc - simulation, xlab = "min",
     ylab = "predicated - measured", cex = 1.5,
     col = "red", main = paste("AIC(Tofts)=",
     map.AIC.T[xx, yy], " AIC(xTofts)=",
     map.AIC.xT[xx, yy], sep = ""))
     lines(locfit(conc - simulation ~ map.times,
     acri = "ici"), col = "red", lwd = 2)
     abline(h = 0, col = "black", lwd = 2)
     if (is.finite(simulation_2[1]) == TRUE) {
     points(map.times, conc - simulation_2,
     xlab = "min", cex = 1.5, col = "blue")
     lines(locfit(conc - simulation_2 ~ map.times,
     acri = "ici"), col = "blue", lwd = 2,
     lty = 2)
     }
     }
     }
     dev.set(4)
     }
     }
     if (file.original == TRUE) {
     ptm <- proc.time()[3]
     proc.time.total <- format((proc.time()[3] - ptm_total)/60,
     digits = 2)
     args <- list(as.character(file), as.character(results_file),
     as.character(method.optimization), show.rt.fits,
     as.character(param.for.avdt), range.map, cutoff.map,
     export.matlab, export.RData, verbose, show.errors,
     try.silent, fracGTzero, AIF.shift, slice, ID.visit,
     est.per.voxel.tlag)
     names(args) <- c("file", "resultsfile", "methodoptimization",
     "showrtfits", "paramforavdt", "rangemap", "cutoffmap",
     "exportmatlab", "exportRData", "verbose", "showerrors",
     "trysilent", "fracGTzero", "AIFshift", "slice",
     "IDvisit", "estpervoxeltlag")
     roiplotparams <- list(x_min, x_max, y_min, y_max,
     MAP_ul)
     names(roiplotparams) <- c("xmin", "xmax", "ymin",
     "ymax", "MAPul")
     dummy_data <- dcemri.data
     dcemri.data <- list(args, map_cc_slice, map_cc_roi,
     cc.median, map.times, aif, aif.shifted, mask.roi,
     map.KtransxT, map.KtransxT.cv, map.tlagxT,
     map.tlagxT.cv, map.kepxT, map.kepxT.cv, map.vbxT,
     map.vbxT.cv, map.vexT, map.OptimValuexT, map.fitfailuresxT,
     param.est.medianxT, roi.median.fitted.xTofts,
     param.est.whole.roi.xTofts, cv.whole.roi.xTofts,
     map.KtransT, map.KtransT.cv, map.kepT, map.kepT.cv,
     map.veT, map.OptimValueT, map.fitfailuresT,
     param.est.medianT, roi.median.fitted.Tofts,
     param.est.whole.roi.Tofts, proc.time.total,
     roiplotparams, KAT.version, map.AIC.xT, map.AIC.T,
     map.AIC.compare, nx, ny, nt, cc_fittedxT, cc_fittedT,
     p0.T, p0.xT, IRF.results, map.EF)
     names(dcemri.data) <- c("args", "cc", "ccroi",
     "ccmedian", "maptimes", "aif", "aifshifted",
     "maskroi", "mapKtransxT", "mapKtransxTcv",
     "maptlagxT", "maptlagxTcv", "mapkepxT", "mapkepxTcv",
     "mapvbxT", "mapvbxTcv", "mapvexT", "mapOptimValuexT",
     "mapfitfailuresxT", "paramestmedianxT", "roimedianfittedxTofts",
     "paramestwholeroixTofts", "cvwholeroixTofts",
     "mapKtransT", "mapKtransTcv", "mapkepT", "mapkepTcv",
     "mapveT", "mapOptimValueT", "mapfitfailuresT",
     "paramestmedianT", "roimedianfittedTofts",
     "paramestwholeroiTofts", "proctimetotal", "roiplotparams",
     "KATversion", "mapAICxT", "mapAICT", "mapAICcompare",
     "nx", "ny", "nt", "ccfittedxT", "ccfittedT",
     "p0T", "p0xT", "IRFresults", "mapEF")
     if (export.RData == TRUE) {
     cat("writing results to ", paste(results_file,
     ".RData", sep = ""), "...", sep = "", "\n")
     save(dcemri.data, file = paste(results_file,
     ".RData", sep = ""))
     }
     if (export.matlab == TRUE) {
     cat("writing results to ", paste(results_file,
     ".mat", sep = ""), "...", sep = "", "\n")
     writeMat(paste(results_file, ".mat", sep = ""),
     args = args, mapccslice = map_cc_slice, mapccroi = map_cc_roi,
     ccmedian = cc.median, maptimes = map.times,
     aif = aif, aifshifted = aif.shifted, maskroi = mask.roi,
     mapKtransxT = map.KtransxT, mapKtransxTcv = map.KtransxT.cv,
     maptlagxT = map.tlagxT, maptlagxTcv = map.tlagxT.cv,
     mapkepxT = map.kepxT, mapkepxTcv = map.kepxT.cv,
     mapvbxT = map.vbxT, mapvbxTcv = map.vbxT.cv,
     mapvexT = map.vexT, mapOptimValuexT = map.OptimValuexT,
     mapfitfailuresxT = map.fitfailuresxT, paramestmedianxT = param.est.medianxT,
     roimedianfittedxTofts = roi.median.fitted.xTofts,
     paramestwholeroixTofts = param.est.whole.roi.xTofts,
     cvwholeroixTofts = cv.whole.roi.xTofts, mapKtransT = map.KtransT,
     mapKtransTcv = map.KtransT.cv, mapkepT = map.kepT,
     mapkepTcv = map.kepT.cv, mapveT = map.veT,
     mapOptimValueT = map.OptimValueT, mapfitfailuresT = map.fitfailuresT,
     paramestmedianT = param.est.medianT, roimedianfittedTofts = roi.median.fitted.Tofts,
     paramestwholeroiTofts = param.est.whole.roi.Tofts,
     proctimetotal = proc.time.total, roiplotparams = roiplotparams,
     KATversion = KAT.version, mapAICxT = map.AIC.xT,
     mapAICT = map.AIC.T, mapAICcompare = map.AIC.compare,
     nx = nx, ny = ny, nt = nt, ccfittedxT = cc_fittedxT,
     ccfittedT = cc_fittedT, p0T = p0.T, p0xT = p0.xT,
     IRFresults = IRF.results, mapEF = map.EF)
     }
     dcemri.data <- dummy_data
     cat("..done in", format((proc.time()[3] - ptm)/60,
     digits = 2), "minutes.", "\n")
     if (export.matlab == FALSE)
     cat("Run KAT(filename.RData) to visualize results.",
     "\n")
     if (export.matlab == TRUE)
     cat("Run KAT(filename.RData) or KAT(filename.mat) to visualize results.",
     "\n")
     cat("--------", "\n")
     }
     }
     }
    }
    <bytecode: 0x491c1d0>
    <environment: namespace:KATforDCEMRI>
     --- function search by body ---
    Function KAT in namespace KATforDCEMRI has this body.
     ----------- END OF FAILURE REPORT --------------
    Error in if (class(cov) != "try-error") { : the condition has length > 1
    Calls: demo ... withVisible -> eval -> eval -> runme -> system.time -> KAT
    Timing stopped at: 1.759 0.139 1.991
    Execution halted
Flavor: r-devel-linux-x86_64-debian-clang

Version: 1.0.1
Check: examples
Result: ERROR
    Running examples in ‘KATforDCEMRI-Ex.R’ failed
    The error most likely occurred in:
    
    > base::assign(".ptime", proc.time(), pos = "CheckExEnv")
    > ### Name: KAT
    > ### Title: Kinetic Analysis Tool for DCE-MRI
    > ### Aliases: KAT
    > ### Keywords: kinetic DCEMRI
    >
    > ### ** Examples
    >
    > ## Create temporary directory for example code output files
    > temp_dir <- tempdir(check=FALSE)
    > ##
    > current_dir <- getwd()
    > setwd(temp_dir)
    > ##
    > ## Run example code
    > demo(KAT, ask=FALSE)
    
    
     demo(KAT)
     ---- ~~~
    
    > ## KATforDCEMRI: a Kinetic Analysis Tool for DCE-MRI
    > ## Copyright 2018 Genentech, Inc.
    > ##
    > ## For questions or comments, please contact
    > ## Gregory Z. Ferl, Ph.D.
    > ## Genentech, Inc.
    > ## Development Sciences
    > ## 1 DNA Way, Mail stop 463A
    > ## South San Francisco, CA, United States of America
    > ## E-mail: ferl.gregory@gene.com
    >
    > runme <- function(){
    + data(dcemri.data, package="KATforDCEMRI")
    +
    + ## dir.create("KATforDCEMRI_benchmark_test")
    + ## setwd("KATforDCEMRI_benchmark_test")
    +
    + attach(dcemri.data)
    +
    + ## SHRINK THE ROI MASK
    + maskROI[,,] <- 0
    + #maskROI[32:42,32:42,] <- 1
    + maskROI[34:36,34:36,] <- 1
    +
    + runtime1 <- system.time(KAT.checkData(file.name="KAT", vector.times=vectorTimes, map.CC=mapCC, mask.ROI=maskROI, vector.AIF=vectorAIF))
    + runtime2 <- system.time(KAT(file = "KAT.RData", results_file="KAT_benchmark_test-full", range.map=1.05, cutoff.map=0.95, AIF.shift="NONE", tlag.Tofts.on=FALSE, export.matlab=FALSE))
    +
    + ## runtime3 <- system.time(KAT.checkData(file.name="KATtrunc", vector.times=vectorTimes[1:44], map.CC=mapCC[,,,1:44], mask.ROI=maskROI, vector.AIF=vectorAIF[1:44]))
    + ## runtime4 <- system.time(KAT(file = "KATtrunc.RData", results_file="KAT_benchmark_test-truncated", range.map=1.05, cutoff.map=0.95))
    + detach(dcemri.data)
    +
    + ## runtime <- format(runtime1[3] + runtime2[3] + runtime3[3] + runtime4[3], digits=3)
    + runtime <- format(runtime1[3] + runtime2[3], digits=3)
    +
    + ## KAT.plot(F1="KAT_benchmark_test-full_slice1.RData", F2="KAT_benchmark_test-full_slice2.RData", F3="KAT_benchmark_test-truncated_slice1.RData", F4="KAT_benchmark_test-truncated_slice2.RData", export.matlab=FALSE)
    + KAT.plot(F1="KAT_benchmark_test-full_slice1.RData", F2="KAT_benchmark_test-full_slice2.RData", F3="KAT_benchmark_test-full_slice3.RData", F4="KAT_benchmark_test-full_slice4.RData", export.matlab=FALSE)
    +
    + load("KAT_benchmark_test-full_slice1.RData")
    + Ktrans_s1A <- as.numeric(format(dcemri.data$paramestmedianxT$Ktrans.median, digits=3))
    + kep_s1A <- as.numeric(format(dcemri.data$paramestmedianxT$kep.median, digits=3))
    + vb_s1A <- as.numeric(format(dcemri.data$paramestmedianxT$vb.median, digits=3))
    + cvKtrans_s1A <- as.numeric(format(100*sd(as.vector(dcemri.data$mapKtransxT), na.rm=TRUE)/Ktrans_s1A, digits=3))
    + cvkep_s1A <- as.numeric(format(100*sd(as.vector(dcemri.data$mapkepxT), na.rm=TRUE)/kep_s1A, digits=3))
    + cvvb_s1A <- as.numeric(format(100*sd(as.vector(dcemri.data$mapvbxT), na.rm=TRUE)/vb_s1A, digits=3))
    +
    + Ktrans_Ts1A <- as.numeric(format(dcemri.data$paramestmedianT$Ktrans.median, digits=3))
    + kep_Ts1A <- as.numeric(format(dcemri.data$paramestmedianT$kep.median, digits=3))
    + cvKtrans_Ts1A <- as.numeric(format(100*sd(as.vector(dcemri.data$mapKtransT), na.rm=TRUE)/Ktrans_Ts1A, digits=3))
    + cvkep_Ts1A <- as.numeric(format(100*sd(as.vector(dcemri.data$mapkepT), na.rm=TRUE)/kep_s1A, digits=3))
    +
    + load("KAT_benchmark_test-full_slice2.RData")
    + Ktrans_s2A <- as.numeric(format(dcemri.data$paramestmedianxT$Ktrans.median, digits=3))
    + kep_s2A <- as.numeric(format(dcemri.data$paramestmedianxT$kep.median, digits=3))
    + vb_s2A <- as.numeric(format(dcemri.data$paramestmedianxT$vb.median, digits=3))
    + cvKtrans_s2A <- as.numeric(format(100*sd(as.vector(dcemri.data$mapKtransxT), na.rm=TRUE)/Ktrans_s2A, digits=3))
    + cvkep_s2A <- as.numeric(format(100*sd(as.vector(dcemri.data$mapkepxT), na.rm=TRUE)/kep_s2A, digits=3))
    + cvvb_s2A <- as.numeric(format(100*sd(as.vector(dcemri.data$mapvbxT), na.rm=TRUE)/vb_s2A, digits=3))
    +
    + Ktrans_Ts2A <- as.numeric(format(dcemri.data$paramestmedianT$Ktrans.median, digits=3))
    + kep_Ts2A <- as.numeric(format(dcemri.data$paramestmedianT$kep.median, digits=3))
    + cvKtrans_Ts2A <- as.numeric(format(100*sd(as.vector(dcemri.data$mapKtransT), na.rm=TRUE)/Ktrans_Ts2A, digits=3))
    + cvkep_Ts2A <- as.numeric(format(100*sd(as.vector(dcemri.data$mapkepT), na.rm=TRUE)/kep_s2A, digits=3))
    +
    + ## load("KAT_benchmark_test-truncated_slice1.RData")
    + load("KAT_benchmark_test-full_slice3.RData")
    + Ktrans_s1B <- as.numeric(format(dcemri.data$paramestmedianxT$Ktrans.median, digits=3))
    + kep_s1B <- as.numeric(format(dcemri.data$paramestmedianxT$kep.median, digits=3))
    + vb_s1B <- as.numeric(format(dcemri.data$paramestmedianxT$vb.median, digits=3))
    + cvKtrans_s1B <- as.numeric(format(100*sd(as.vector(dcemri.data$mapKtransxT), na.rm=TRUE)/Ktrans_s1B, digits=3))
    + cvkep_s1B <- as.numeric(format(100*sd(as.vector(dcemri.data$mapkepxT), na.rm=TRUE)/kep_s1B, digits=3))
    + cvvb_s1B <- as.numeric(format(100*sd(as.vector(dcemri.data$mapvbxT), na.rm=TRUE)/vb_s1B, digits=3))
    +
    + Ktrans_Ts1B <- as.numeric(format(dcemri.data$paramestmedianT$Ktrans.median, digits=3))
    + kep_Ts1B <- as.numeric(format(dcemri.data$paramestmedianT$kep.median, digits=3))
    + cvKtrans_Ts1B <- as.numeric(format(100*sd(as.vector(dcemri.data$mapKtransT), na.rm=TRUE)/Ktrans_Ts1B, digits=3))
    + cvkep_Ts1B <- as.numeric(format(100*sd(as.vector(dcemri.data$mapkepT), na.rm=TRUE)/kep_s1B, digits=3))
    +
    + ## load("KAT_benchmark_test-truncated_slice2.RData")
    + load("KAT_benchmark_test-full_slice4.RData")
    + Ktrans_s2B <- as.numeric(format(dcemri.data$paramestmedianxT$Ktrans.median, digits=3))
    + kep_s2B <- as.numeric(format(dcemri.data$paramestmedianxT$kep.median, digits=3))
    + vb_s2B <- as.numeric(format(dcemri.data$paramestmedianxT$vb.median, digits=3))
    + cvKtrans_s2B <- as.numeric(format(100*sd(as.vector(dcemri.data$mapKtransxT), na.rm=TRUE)/Ktrans_s2B, digits=3))
    + cvkep_s2B <- as.numeric(format(100*sd(as.vector(dcemri.data$mapkepxT), na.rm=TRUE)/kep_s2B, digits=3))
    + cvvb_s2B <- as.numeric(format(100*sd(as.vector(dcemri.data$mapvbxT), na.rm=TRUE)/vb_s2B, digits=3))
    +
    + Ktrans_Ts2B <- as.numeric(format(dcemri.data$paramestmedianT$Ktrans.median, digits=3))
    + kep_Ts2B <- as.numeric(format(dcemri.data$paramestmedianT$kep.median, digits=3))
    + cvKtrans_Ts2B <- as.numeric(format(100*sd(as.vector(dcemri.data$mapKtransT), na.rm=TRUE)/Ktrans_Ts2B, digits=3))
    + cvkep_Ts2B <- as.numeric(format(100*sd(as.vector(dcemri.data$mapkepT), na.rm=TRUE)/kep_s2B, digits=3))
    +
    + pdf(file="KAT_demo-page1.pdf", height=11, width=8.5)
    +
    + plot(0:35, type = "n", xaxt="n", yaxt="n", bty="n", xlab = "", ylab = "")
    + text(5, 35, paste("KATforDCEMRI version ", dcemri.data$KATversion, " BENCHMARK TEST", sep=""), pos=4, font=2, col="red")
    + text(5, 33, paste("date:", date(), "\n"), pos=4)
    + text(5, 32, paste("Total Processing Time:", runtime, "seconds"), pos=4)
    + text(5, 29, paste("sysname/release:", Sys.info()[[1]], Sys.info()[[2]], "\n"), pos=4)
    + text(5, 28, paste("nodename:", Sys.info()[[4]], "\n"), pos=4)
    + text(5, 27, paste("user:", Sys.info()[[7]], "\n"), pos=4)
    + text(5, 25, "Estimated (inter-voxel %CV) versus True Parameter values (extended Tofts)", pos=4, font=2)
    + text(5, 23, "LOW NOISE DATA", pos=4)
    + text(5, 22, paste("Slice 1: Ktrans =", Ktrans_s1A, "1/min (", cvKtrans_s1A,"%) [true value = 0.22 1/min]", sep=""), pos=4)
    + text(5, 21, paste("Slice 1: kep =", kep_s1A, "1/min (", cvkep_s1A, "%) [true value = 1.1 1/min]", sep=""), pos=4)
    + text(5, 20, paste("Slice 1: vb =", vb_s1A, " (", cvvb_s1A, "%) [true value = 0]", sep=""), pos=4)
    + text(5, 18, paste("Slice 2: Ktrans =", Ktrans_s2A, "1/min (", cvKtrans_s2A,"%) [true value = 0.22 1/min]", sep=""), pos=4)
    + text(5, 17, paste("Slice 2: kep =", kep_s2A, "1/min (", cvkep_s2A, "%) [true value = 1.1 1/min]", sep=""), pos=4)
    + text(5, 16, paste("Slice 2: vb =", vb_s2A, " (", cvvb_s2A, "%) [true value = 0.05]", sep=""), pos=4)
    + text(5, 14, "HIGH NOISE DATA", pos=4)
    + text(5, 13, paste("Slice 3: Ktrans =", Ktrans_s1B, "1/min (", cvKtrans_s1B,"%) [true value = 0.22 1/min]", sep=""), pos=4)
    + text(5, 12, paste("Slice 3: kep =", kep_s1B, "1/min (", cvkep_s1B, "%) [true value = 1.1 1/min]", sep=""), pos=4)
    + text(5, 11, paste("Slice 3: vb =", vb_s1B, " (", cvvb_s1B, "%) [true value = 0]", sep=""), pos=4)
    + text(5, 9, paste("Slice 4: Ktrans =", Ktrans_s2B, "1/min (", cvKtrans_s2B,"%) [true value = 0.22 1/min]", sep=""), pos=4)
    + text(5, 8, paste("Slice 4: kep =", kep_s2B, "1/min (", cvkep_s2B, "%) [true value = 1.1 1/min]", sep=""), pos=4)
    + text(5, 7, paste("Slice 4: vb =", vb_s2B, " (", cvvb_s2B, "%) [true value = 0.05]", sep=""), pos=4)
    + dev.off()
    +
    +
    + pdf(file="KAT_demo-page2.pdf", height=11, width=8.5)
    +
    + plot(0:35, type = "n", xaxt="n", yaxt="n", bty="n", xlab = "", ylab = "")
    +
    + text(5, 25, "Estimated (inter-voxel %CV) versus True Parameter values (Tofts Model)", pos=4, font=2)
    + text(5, 23, "LOW NOISE DATA", pos=4)
    + text(5, 22, paste("Slice 1: Ktrans =", Ktrans_Ts1A, "1/min (", cvKtrans_Ts1A,"%) [true value = 0.22 1/min]", sep=""), pos=4)
    + text(5, 21, paste("Slice 1: kep =", kep_Ts1A, "1/min (", cvkep_Ts1A, "%) [true value = 1.1 1/min]", sep=""), pos=4)
    + text(5, 20, "Slice 1: vb = 0 (Fixed) [true value = 0]", pos=4)
    + text(5, 18, paste("Slice 2: Ktrans =", Ktrans_Ts2A, "1/min (", cvKtrans_Ts2A,"%) [true value = 0.22 1/min]", sep=""), pos=4)
    + text(5, 17, paste("Slice 2: kep =", kep_Ts2A, "1/min (", cvkep_Ts2A, "%) [true value = 1.1 1/min]", sep=""), pos=4)
    + text(5, 16, "Slice 2: vb = 0 (Fixed) [true value = 0.05]", pos=4)
    + text(5, 14, "HIGH NOISE DATA", pos=4)
    + text(5, 13, paste("Slice 3: Ktrans =", Ktrans_Ts1B, "1/min (", cvKtrans_Ts1B,"%) [true value = 0.22 1/min]", sep=""), pos=4)
    + text(5, 12, paste("Slice 3: kep =", kep_Ts1B, "1/min (", cvkep_Ts1B, "%) [true value = 1.1 1/min]", sep=""), pos=4)
    + text(5, 11, "Slice 3: vb = 0 (Fixed) [true value = 0]", pos=4)
    + text(5, 9, paste("Slice 4: Ktrans =", Ktrans_Ts2B, "1/min (", cvKtrans_Ts2B,"%) [true value = 0.22 1/min]", sep=""), pos=4)
    + text(5, 8, paste("Slice 4: kep =", kep_Ts2B, "1/min (", cvkep_Ts2B, "%) [true value = 1.1 1/min]", sep=""), pos=4)
    + text(5, 7, "Slice 4: vb = 0 (Fixed) [true value = 0.05]", pos=4)
    + dev.off()
    + }
    
    > runme()
    
    checking dimensions of vectors and arrays...
    
    length of vector.times is 89 with units of seconds
    length of vector.AIF is 89
    dimensions of map.CC array are 75 x 75 x 4 slices x 89 time points
    dimensions of mask.ROI array are 75 x 75 x 4 slices
    
    ...vector and array dimensions are okay.
    
    Saving data in a single R file...
    ...file saved as KAT.RData ...
    ...use the KAT() function to analyze data within this file.
    
    
    #########################################################################
    ##-------------- KINETIC ANALYSIS TOOL (KAT) FOR DCE-MRI --------------##
    #########################################################################
    ##---------------------- R package version 1.0 ----------------------##
    #########################################################################
    
    loading KAT.RData into R...
    ..done in 0.001 minutes.
    --------
    ***** ROI DETECTED IN SLICE 1 *****
    --------
    extracting slice 1 for analysis...
    ..done in 0.0015 minutes.
    --------
    applying ROI mask to cc matrix...
    ..done in 0.00028 minutes.
    --------
    fitting xTofts and Tofts models to whole ROI data...
    ..done in 0.0024 minutes.
    --------
     ----------- FAILURE REPORT --------------
     --- failure: the condition has length > 1 ---
     --- srcref ---
    :
     --- package (from environment) ---
    KATforDCEMRI
     --- call from context ---
    KAT(file = "KAT.RData", results_file = "KAT_benchmark_test-full",
     range.map = 1.05, cutoff.map = 0.95, AIF.shift = "NONE",
     tlag.Tofts.on = FALSE, export.matlab = FALSE)
     --- call from argument ---
    if (class(cov) != "try-error") {
     sd <- try(sqrt(diag(cov)), silent = try.silent)
     if (class(sd) != "try-error") {
     cv <- sd/param_est * 100
     cv.whole.roi.xTofts <- as.numeric(format(cv, digits = 1))
     }
    }
     --- R stacktrace ---
    where 1 at /home/hornik/tmp/R.check/r-devel-gcc/Work/build/Packages/KATforDCEMRI/demo/KAT.R#26: KAT(file = "KAT.RData", results_file = "KAT_benchmark_test-full",
     range.map = 1.05, cutoff.map = 0.95, AIF.shift = "NONE",
     tlag.Tofts.on = FALSE, export.matlab = FALSE)
    where 2 at /home/hornik/tmp/R.check/r-devel-gcc/Work/build/Packages/KATforDCEMRI/demo/KAT.R#26: system.time(KAT(file = "KAT.RData", results_file = "KAT_benchmark_test-full",
     range.map = 1.05, cutoff.map = 0.95, AIF.shift = "NONE",
     tlag.Tofts.on = FALSE, export.matlab = FALSE))
    where 3 at /home/hornik/tmp/R.check/r-devel-gcc/Work/build/Packages/KATforDCEMRI/demo/KAT.R#141: runme()
    where 4: eval(ei, envir)
    where 5: eval(ei, envir)
    where 6: withVisible(eval(ei, envir))
    where 7: source(available, echo = echo, max.deparse.length = Inf, keep.source = TRUE,
     encoding = encoding)
    where 8: demo(KAT, ask = FALSE)
    
     --- value of length: 2 type: logical ---
    [1] TRUE TRUE
     --- function from context ---
    function (file = "concatenate.KAT.with.KAT.checkData.RData",
     results_file = "my_results", method.optimization = "L-BFGS-B",
     show.rt.fits = FALSE, param.for.avdt = "Ktrans", range.map = 1.5,
     cutoff.map = 0.85, export.matlab = TRUE, export.RData = TRUE,
     verbose = FALSE, show.errors = FALSE, try.silent = TRUE,
     fracGTzero = 0.75, AIF.shift = "", Force.AIF.peak = FALSE,
     tlag.Tofts.on = FALSE, est.per.voxel.tlag = FALSE, ...)
    {
     lo <- 0
     options(show.error.messsages = show.errors)
     ftype <- strsplit(file, split = "a")[[1]]
     ftype <- ftype[length(ftype) - 1]
     ftype <- strsplit(ftype, split = "")[[1]]
     ftype <- ftype[length(ftype)]
     if (ftype == "m")
     file.format <- "matlab"
     if (ftype == "D")
     file.format <- "RData"
     file_short <- file
     KAT.version <- "1.0"
     ptm_total <- proc.time()[3]
     modeltype1 <- "xTofts"
     modeltype2 <- "Tofts"
     funcz <- function(x) {
     as.numeric(as.character(x))
     }
     aif.shift.func <- function(t, cp, time_shift) {
     x <- cbind(t, cp)
     tmax <- subset(x[, 1], x[, 2] == max(x[, 2]))
     if (AIF.shift == "ARTERY")
     tmax <- tmax + time_shift
     if (AIF.shift == "VEIN")
     tmax <- tmax - time_shift
     cpFUNC <- approxfun(t, cp, rule = 2)
     if (AIF.shift == "ARTERY")
     tshift <- t - time_shift
     if (AIF.shift == "VEIN")
     tshift <- t + time_shift
     cp.shift <- cpFUNC(tshift)
     if (Force.AIF.peak == TRUE)
     cp.shift[cp.shift == max(cp.shift)] <- max(cp)
     return(cp.shift)
     }
     roi.modelT <- function(p, t, dt, cp, zing = 0) {
     if (zing == 0) {
     Ktrans <- p[1]
     kep <- p[2]
     if (tlag.Tofts.on == TRUE) {
     if (fix.tlag != TRUE & AIF.shift != "NONE")
     time_shift <- p[3]
     if (fix.tlag == TRUE & AIF.shift != "NONE")
     time_shift <- param.est.whole.roi.Tofts$tlag
     }
     if (tlag.Tofts.on == FALSE) {
     time_shift <- param.est.whole.roi.xTofts$tlag
     }
     if (AIF.shift == "ARTERY" | AIF.shift == "VEIN")
     cp <- aif.shift.func(t, cp, time_shift)
     f <- Ktrans * exp(kep * t) * cp
     int <- c(0, cumsum(dt * (tail(f, -1) + head(f, -1)))/2)
     ct <- exp(-kep * t) * int
     return(ct)
     }
     if (zing == 1)
     return("modelT")
     }
     roi.modelxT <- function(p, t, dt, cp, zing = 0) {
     if (zing == 0) {
     Ktrans <- p[1]
     kep <- p[2]
     vb <- p[3]
     if (fix.tlag != TRUE & AIF.shift != "NONE")
     time_shift <- p[4]
     if (est.per.voxel.tlag == FALSE) {
     if (fix.tlag == TRUE & AIF.shift != "NONE")
     time_shift <- param.est.whole.roi.xTofts$tlag
     }
     if (est.per.voxel.tlag == TRUE) {
     if (fix.tlag == TRUE & AIF.shift != "NONE")
     time_shift <- p[4]
     }
     if (AIF.shift == "ARTERY" | AIF.shift == "VEIN")
     cp <- aif.shift.func(t, cp, time_shift)
     f <- Ktrans * exp(kep * t) * cp
     int <- c(0, cumsum(dt * (tail(f, -1) + head(f, -1)))/2)
     ct <- exp(-kep * t) * int
     ct <- ct + vb * cp
     return(ct)
     }
     if (zing == 1)
     return("modelxT")
     }
     calch <- function(u, y, TIME_trunc) {
     locfit_y <- preplot(y, newdata = 0:max(TIME_trunc))
     y_smooth <- locfit_y$fit
     u <- u[match(u[u == max(u)], u):length(u)]
     y_smooth <- y_smooth[match(u[u == max(u)], u):length(u)]
     n <- length(u)
     A <- matrix(0, nrow = n, ncol = n)
     ind <- row(A) - col(A)
     ind[ind < 0] <- (-1)
     ind <- ind + 2
     A <- matrix(c(0, u)[ind], nrow = n, ncol = n)
     h <- solve(A, y_smooth)
     h.time.vector <- (1:length(h)) - 1
     out <- list(h.time.vector, h)
     names(out) <- c("t", "IRF")
     return(out)
     }
     calchFUNC <- function(vector.times, AIF, map_cc_slice, correct.vp = TRUE,
     alpha.AIF = c(0.1, 0.5), vp.nom = 0.1, kep.nom = 0.5) {
     AUMC <- function(AUMC.median, h.median, irf_time_vec,
     r) {
     AUMC.median <- AUMC.median + 0.5 * (h.median[r] *
     irf_time_vec[r] + h.median[r + 1] * irf_time_vec[r +
     1])
     }
     artery_data <- data.frame(vector.times * 60, AIF)
     names(artery_data) <- c("TIME", "ARTERY")
     data_artery_peak <- subset(artery_data, artery_data$ARTERY ==
     max(artery_data$ARTERY))
     data_remove_artery_prepeak <- subset(artery_data, artery_data$TIME >=
     data_artery_peak$TIME)
     frames_to_peak <- length(artery_data[, 1]) - length(data_remove_artery_prepeak[,
     1]) + 1
     TIME <- data_remove_artery_prepeak$TIME
     ARTERY <- data_remove_artery_prepeak$ARTERY
     TIME_trunc <- TIME[seq(1, length(TIME) - 1, by = 1)]
     TIME_trunc <- TIME_trunc - TIME_trunc[1]
     ARTERY_trunc <- ARTERY[seq(1, length(ARTERY) - 1, by = 1)]
     ARTERY_smooth <- locfit.robust(ARTERY_trunc ~ TIME_trunc,
     acri = "cp", alpha = alpha.AIF)
     AIF_smooth <- ARTERY_smooth
     locfit_u <- preplot(AIF_smooth, newdata = 0:max(TIME_trunc))
     u_smooth <- locfit_u$fit
     Tmax <- max(TIME_trunc)
     TUMOR.median <- map_cc_slice
     TUMOR.median <- TUMOR.median[seq(frames_to_peak, length(TUMOR.median),
     by = 1)]
     if (vp.nom > 0)
     TUMOR.median_corr <- TUMOR.median - vp.nom * ARTERY
     if (vp.nom <= 0)
     TUMOR.median_corr <- TUMOR.median
     TUMOR.median_corr_shifted <- TUMOR.median_corr[seq(2,
     length(TUMOR.median_corr), by = 1)]
     TUMOR.median_smooth <- locfit.robust(TUMOR.median_corr_shifted ~
     TIME_trunc, acri = "cp")
     calch.out <- calch(u_smooth, TUMOR.median_smooth, TIME_trunc)
     h.median <- calch.out$IRF
     irf_time_vec <- calch.out$t
     n <- length(h.median)
     AUC.median <- 0
     AUMC.median <- 0
     for (r in 1:(n - 1)) {
     h_sum <- h.median[r] + h.median[r + 1]
     t_sum <- irf_time_vec[r] + irf_time_vec[r + 1]
     AUC.median <- AUC.median + 0.5 * h_sum
     AUMC.median <- AUMC(AUMC.median, h.median, irf_time_vec,
     r)
     }
     AUCMRT.median <- AUC.median/(AUMC.median/AUC.median) *
     60
     if (kep.nom > 0) {
     t_scan <- max(TIME_trunc)/60
     ve_trunc_error <- 1 - exp(-kep.nom * t_scan)
     Ktrans_trunc_error <- (1 - exp(-kep.nom * t_scan))^2/(1 -
     (1 + kep.nom * t_scan) * exp(-kep.nom * t_scan))
     AUC.median <- AUC.median/ve_trunc_error
     AUCMRT.median <- AUCMRT.median/Ktrans_trunc_error
     }
     irf_time_vec <- irf_time_vec/60
     out <- list(AUC.median, AUCMRT.median, h.median, irf_time_vec)
     names(out) <- c("AUCh", "AUChMRTh", "IRF", "t")
     return(out)
     }
     Obj_roi <- function(p, model, t, dt, cp, roi) {
     sum((model(p, t, dt, cp) - roi)^2)
     }
     map_cc_slice <- NULL
     map_cc_roi <- NULL
     aif <- NULL
     aif.shifted <- NULL
     map.times <- NULL
     map.KtransxT <- NULL
     map.kepxT <- NULL
     map.vexT <- NULL
     map.vbxT <- NULL
     map.tlagxT <- NULL
     map.fitfailuresxT <- NULL
     map.KtransT <- NULL
     map.kepT <- NULL
     map.veT <- NULL
     map.fitfailuresT <- NULL
     mask.roi <- NULL
     nx <- NULL
     ny <- NULL
     nt <- NULL
     map.KtransxT <- NULL
     map.kepxT <- NULL
     map.vexT <- NULL
     map.vbxT <- NULL
     map.AIC.compare <- NULL
     map.AIC.T <- NULL
     map.EF <- NULL
     map.AIC.xT <- NULL
     roi.median.fitted.Tofts <- NULL
     param.est.whole.roi.Tofts <- NULL
     roi.median.fitted.xTofts <- NULL
     param.est.whole.roi.xTofts <- NULL
     cv.whole.roi.xTofts <- NULL
     cat("\n")
     cat("#########################################################################",
     "\n")
     cat("##-------------- KINETIC ANALYSIS TOOL (KAT) FOR DCE-MRI --------------##",
     "\n")
     cat("#########################################################################",
     "\n")
     cat("##---------------------- R package version", KAT.version,
     "----------------------##", "\n")
     cat("#########################################################################",
     "\n")
     cat("\n")
     filea <- strsplit(file, split = "/")[[1]]
     fileb <- filea[length(filea)]
     if (file != "concatenate.KAT.with.KAT.checkData.RData")
     cat("loading", fileb, "into R...", "\n")
     ptm <- proc.time()[3]
     if (file.format == "matlab") {
     mat_data <- readMat(file)
     if (length(names(mat_data)) < 10)
     file.original <- TRUE
     if (length(names(mat_data)) >= 10)
     file.original <- FALSE
     }
     if (file.format == "RData") {
     if (file != "concatenate.KAT.with.KAT.checkData.RData")
     load(file)
     if (length(names(dcemri.data)) < 10) {
     file.original <- TRUE
     ROIcounter <- apply(dcemri.data$maskROI, 3, max)
     results_file_temp <- results_file
     }
     if (length(names(dcemri.data)) >= 10) {
     file.original <- FALSE
     ROIcounter <- 1
     }
     }
     if (file != "concatenate.KAT.with.KAT.checkData.RData") {
     cat("..done in", format((proc.time()[3] - ptm)/60, digits = 2),
     "minutes.", "\n")
     cat("--------", "\n")
     }
     for (slicenumber in 1:(length(ROIcounter))) {
     if (ROIcounter[slicenumber] == 1) {
     if (file.original == TRUE) {
     slice <- slicenumber
     cat("***** ROI DETECTED IN SLICE", slice, " *****",
     "\n")
     cat("--------", "\n")
     }
     if (file.original == TRUE) {
     if (AIF.shift != "VEIN" & AIF.shift != "ARTERY" &
     AIF.shift != "NONE")
     stop("You must specify the argument AIF.shift argument as VEIN, ARTERY or NONE, indicating that the AIF you are using is based on data from a vein or artery or NONE if tlag should be set to 0. This will ensure that the time lag parameter in the Tofts and xTofts models has the appropriate inital value and is bounded correctly; either -Inf to 0 (for VEIN) or 0 to Inf (for ARTERY).")
     filenameTag <- paste("_slice", slice, sep = "")
     results_file <- paste(results_file_temp, filenameTag,
     sep = "")
     roi.model <- roi.modelxT
     if (slice == "" || slice < 0)
     stop("The slice argument has not been properly specified; slice=``slice number''")
     cat("extracting slice", slice, "for analysis...",
     "\n")
     ptm <- proc.time()[3]
     if (file.format == "matlab") {
     map.times <- as.vector(mat_data$map[[4]]/60)
     map_cc <- mat_data$map[[3]]
     map_cc_slice <- map_cc[, , slice, ]
     if (length(map.times) != length(map_cc_slice[1,
     1, ]))
     stop("The length of the time vector does not match the length of the contrast agent concentration vector")
     }
     if (file.format == "RData") {
     map.times <- as.vector(dcemri.data$vectorTimes/60)
     map_cc <- dcemri.data$mapCC
     map_cc_slice <- map_cc[, , slice, ]
     if (length(map.times) != length(map_cc_slice[1,
     1, ]))
     stop("The length of the time vector does not match the length of the contrast agent concentration vector")
     }
     nt <- length(map_cc_slice[1, 1, ])
     ny <- length(map_cc_slice[, 1, 1])
     nx <- length(map_cc_slice[1, , 1])
     ccTEMP <- rep(0, prod(dim(map_cc_slice)))
     dim(ccTEMP) <- dim(map_cc_slice)[c(2, 1, 3)]
     for (i in 1:nt) ccTEMP[, , i] <- rot90(map_cc_slice[,
     , i], 3)
     map_cc_slice <- ccTEMP
     if (file.format == "matlab")
     mask.roi <- mat_data$mask[[1]][, , slice]
     if (file.format == "RData")
     mask.roi <- dcemri.data$mask[, , slice]
     mask.roi <- rot90(mask.roi, 3)
     if (max(mask.roi) != 1)
     stop("Your ROI mask is either composed entirely of zeroes or contains nonnumeric elements; voxels within the ROI should have a value of ``1'' and all other voxels should have a value of ``0''.")
     if (file.format == "matlab")
     aif <- as.vector(mat_data$aif)
     if (file.format == "RData")
     aif <- as.vector(dcemri.data$vectorAIF)
     if (file.format == "matlab")
     rm(mat_data)
     cat("..done in", format((proc.time()[3] - ptm)/60,
     digits = 2), "minutes.", "\n")
     cat("--------", "\n")
     cat("applying ROI mask to cc matrix...", "\n")
     ptm <- proc.time()[3]
     map_cc_roi <- map_cc_slice
     mask.roi.temp <- mask.roi
     mask.roi.temp[mask.roi.temp == 0] <- NA
     for (z in 1:nt) map_cc_roi[, , z] <- map_cc_roi[,
     , z] * mask.roi.temp
     cat("..done in", format((proc.time()[3] - ptm)/60,
     digits = 2), "minutes.", "\n")
     cat("--------", "\n")
     ptm <- proc.time()[3]
     cc.median <- seq(1:nt)
     for (t in 1:nt) cc.median[t] <- median(map_cc_roi[,
     , t], na.rm = TRUE)
     cat("fitting", modeltype1, "and", modeltype2,
     "models to whole ROI data...", "\n")
     IRF.out <- calchFUNC(map.times, aif, cc.median)
     AUCcorrnom <- IRF.out$AUCh
     AUCMRTcorrnom <- IRF.out$AUChMRTh
     AUCMRTcorrnom.divby.AUCcorrnom <- IRF.out$AUChMRTh/IRF.out$AUCh
     if (tlag.Tofts.on == TRUE & AIF.shift == "VEIN") {
     p0.T.median <- c(IRF.out$AUChMRTh, IRF.out$AUChMRTh/IRF.out$AUCh,
     0.05)
     lower.wholeT <- c(lo, lo, 0)
     upper.wholeT <- c(Inf, Inf, Inf)
     }
     if (tlag.Tofts.on == TRUE & AIF.shift == "ARTERY") {
     p0.T.median <- c(IRF.out$AUChMRTh, IRF.out$AUChMRTh/IRF.out$AUCh,
     0.05)
     lower.wholeT <- c(lo, lo, 0)
     upper.wholeT <- c(Inf, Inf, Inf)
     }
     if (tlag.Tofts.on == FALSE) {
     p0.T.median <- c(IRF.out$AUChMRTh, IRF.out$AUChMRTh/IRF.out$AUCh)
     lower.wholeT <- c(lo, lo)
     upper.wholeT <- c(Inf, Inf)
     }
     if (AIF.shift == "VEIN") {
     p0.xT.median <- c(IRF.out$AUChMRTh, IRF.out$AUChMRTh/IRF.out$AUCh,
     0.05, 0.05)
     lower.wholexT <- c(lo, lo, lo, 0)
     upper.wholexT <- c(Inf, Inf, Inf, Inf)
     }
     if (AIF.shift == "ARTERY") {
     p0.xT.median <- c(IRF.out$AUChMRTh, IRF.out$AUChMRTh/IRF.out$AUCh,
     0.05, 0.05)
     lower.wholexT <- c(lo, lo, lo, 0)
     upper.wholexT <- c(Inf, Inf, Inf, Inf)
     }
     if (AIF.shift == "NONE") {
     p0.xT.median <- c(IRF.out$AUChMRTh, IRF.out$AUChMRTh/IRF.out$AUCh,
     0.05)
     lower.wholexT <- c(lo, lo, lo)
     upper.wholexT <- c(Inf, Inf, Inf)
     }
     SAAMII <- FALSE
     if (SAAMII == TRUE) {
     hw.x <- cbind(format(map.times, digits = 3),
     format(aif, digits = 3), format(cc.median,
     digits = 3))
     write.table("DATA", file = paste("slice", slice,
     "_forSAAMII", sep = ""), quote = FALSE, row.names = FALSE,
     col.names = FALSE)
     write.table("(SD 1)", file = paste("slice",
     slice, "_forSAAMII", sep = ""), quote = FALSE,
     row.names = FALSE, append = TRUE, col.names = FALSE)
     write.table(hw.x, file = paste("slice", slice,
     "_forSAAMII", sep = ""), quote = FALSE, row.names = FALSE,
     append = TRUE, col.names = c("t", paste("AIF_s",
     slice, sep = ""), paste("CC_s", slice,
     sep = "")))
     write.table("END", file = paste("slice", slice,
     "_forSAAMII", sep = ""), quote = FALSE, row.names = FALSE,
     append = TRUE, col.names = FALSE)
     }
     roi <- cc.median
     t <- map.times
     fix.tlag <- FALSE
     roi.model <- roi.modelxT
     if (method.optimization == "L-BFGS-B")
     fit.roi.median.xTofts <- try(optim(p0.xT.median,
     Obj_roi, model = roi.model, t = map.times,
     dt = diff(map.times), cp = aif, roi = roi,
     method = "L-BFGS-B", lower = lower.wholexT,
     upper = upper.wholexT, hessian = TRUE), silent = try.silent)
     if (method.optimization != "L-BFGS-B")
     fit.roi.median.xTofts <- try(optim(p0.xT.median,
     Obj_roi, model = roi.model, t = map.times,
     dt = diff(map.times), cp = aif, roi = roi,
     method = method.optimization, hessian = TRUE),
     silent = try.silent)
     if (class(fit.roi.median.xTofts) != "try-error") {
     if (fit.roi.median.xTofts$convergence == 0) {
     roi.median.fitted.xTofts <- roi.model(p = fit.roi.median.xTofts$par,
     t = map.times, dt = diff(map.times), cp = aif)
     params.xTofts <- fit.roi.median.xTofts$par
     param.est.whole.roi.xTofts <- list(fit.roi.median.xTofts$par[1],
     fit.roi.median.xTofts$par[2], fit.roi.median.xTofts$par[3],
     fit.roi.median.xTofts$par[4])
     names(param.est.whole.roi.xTofts) <- c("Ktrans",
     "kep", "vb", "tlag")
     }
     }
     if (class(fit.roi.median.xTofts) == "try-error") {
     cat("A problem occured when fitting the extended Tofts model to median intensity/concentration data across the ROI. Optimization algorithm did not converge. This may occur when a large fraction of voxels within the ROI are non-enhancing, as these are not excluded from analysis of the median intensity/concentration profile",
     "\n")
     }
     if (class(fit.roi.median.xTofts) != "try-error") {
     if (fit.roi.median.xTofts$convergence != 0) {
     cat("A problem occured when fitting the Tofts model to median intensity/concentration data across the ROI (convergence code not equal to zero). Optimization algorithm did not converge. This may occur when a large fraction of voxels within the ROI are non-enhancing, as these are not excluded from analysis of the median intensity/concentration profile",
     "\n")
     }
     }
     roi.model <- roi.modelT
     if (method.optimization == "L-BFGS-B")
     fit.roi.median.Tofts <- try(optim(p0.T.median,
     Obj_roi, model = roi.model, t = map.times,
     dt = diff(map.times), cp = aif, roi = roi,
     method = "L-BFGS-B", lower = lower.wholeT,
     upper = upper.wholeT, hessian = TRUE), silent = try.silent)
     if (method.optimization != "L-BFGS-B")
     fit.roi.median.Tofts <- try(optim(p0.T.median,
     Obj_roi, model = roi.model, t = map.times,
     dt = diff(map.times), cp = aif, roi = roi,
     method = method.optimization, hessian = TRUE),
     silent = try.silent)
     if (class(fit.roi.median.Tofts) != "try-error") {
     if (fit.roi.median.Tofts$convergence == 0) {
     roi.median.fitted.Tofts <- roi.model(p = fit.roi.median.Tofts$par,
     t = map.times, dt = diff(map.times), cp = aif)
     if (tlag.Tofts.on == FALSE) {
     param.est.whole.roi.Tofts <- list(fit.roi.median.Tofts$par[1],
     fit.roi.median.Tofts$par[2])
     names(param.est.whole.roi.Tofts) <- c("Ktrans",
     "kep")
     }
     if (tlag.Tofts.on == TRUE) {
     param.est.whole.roi.Tofts <- list(fit.roi.median.Tofts$par[1],
     fit.roi.median.Tofts$par[2], fit.roi.median.Tofts$par[3])
     names(param.est.whole.roi.Tofts) <- c("Ktrans",
     "kep", "tlag")
     }
     }
     }
     if (class(fit.roi.median.Tofts) == "try-error") {
     cat("A problem occured when fitting the Tofts model to median intensity/concentration data across the ROI (a try-error occured). Optimization algorithm did not converge. This may occur when a large fraction of voxels within the ROI are non-enhancing, as these are not excluded from analysis of the median intensity/concentration profile",
     "\n")
     }
     if (class(fit.roi.median.Tofts) != "try-error") {
     if (fit.roi.median.Tofts$convergence != 0) {
     cat("A problem occured when fitting the Tofts model to median intensity/concentration data across the ROI (convergence code not equal to zero). Optimization algorithm did not converge. This may occur when a large fraction of voxels within the ROI are non-enhancing, as these are not excluded from analysis of the median intensity/concentration profile",
     "\n")
     }
     }
     roi.model <- roi.modelxT
     IRF.out <- calchFUNC(vector.times = map.times,
     AIF = aif, map_cc_slice = cc.median, vp.nom = param.est.whole.roi.xTofts$vb,
     kep.nom = param.est.whole.roi.xTofts$kep)
     p0.T <- c(IRF.out$AUChMRTh, IRF.out$AUChMRTh/IRF.out$AUCh)
     names(p0.T) <- c("Ktrans", "kep")
     if (est.per.voxel.tlag == FALSE) {
     p0.xT <- c(IRF.out$AUChMRTh, IRF.out$AUChMRTh/IRF.out$AUCh,
     param.est.whole.roi.xTofts$vb)
     names(p0.xT) <- c("Ktrans", "kep", "vb")
     }
     if (est.per.voxel.tlag == TRUE) {
     p0.xT <- c(IRF.out$AUChMRTh, IRF.out$AUChMRTh/IRF.out$AUCh,
     param.est.whole.roi.xTofts$vb, 0.05)
     names(p0.xT) <- c("Ktrans", "kep", "vb", "tlag")
     }
     AUCcorr <- IRF.out$AUCh
     AUCMRTcorr <- IRF.out$AUChMRTh
     AUCMRTcorr.divby.AUCcorr <- IRF.out$AUChMRTh/IRF.out$AUCh
     IRF.results <- c(AUCcorrnom, AUCMRTcorrnom, AUCMRTcorrnom.divby.AUCcorrnom,
     AUCcorr, AUCMRTcorr, AUCMRTcorr.divby.AUCcorr)
     names(IRF.results) <- c("AUCcorrnom(ve)", "AUCMRTcorrnom(Ktrans)",
     "AUCMRTcorrnom.divby.AUCcorrnom(kep)", "AUCcorr(ve)",
     "AUCMRTcorr(Ktrans)", "AUCMRTcorr.divby.AUCcorr(kep)")
     cat("..done in", format((proc.time()[3] - ptm)/60,
     digits = 2), "minutes.", "\n")
     cat("--------", "\n")
     roi.model <- roi.modelxT
     if (class(fit.roi.median.xTofts) != "try-error") {
     if (fit.roi.median.xTofts$convergence == 0) {
     RSS <- sum((roi.model(p = fit.roi.median.xTofts$par,
     t = map.times, dt = diff(map.times), cp = aif) -
     roi)^2)
     param_est <- fit.roi.median.xTofts$par
     nD <- nt
     nP <- length(param_est)
     hess <- fit.roi.median.xTofts$hessian
     cov <- try((nP * RSS/(nD - nP)) * solve(hess),
     silent = try.silent)
     cv <- param_est
     cv[] <- NA
     if (class(cov) != "try-error") {
     sd <- try(sqrt(diag(cov)), silent = try.silent)
     if (class(sd) != "try-error") {
     cv <- sd/param_est * 100
     cv.whole.roi.xTofts <- as.numeric(format(cv,
     digits = 1))
     }
     }
     param.roi <- as.numeric(format(param_est,
     digits = 3))
     }
     }
     if (AIF.shift == "ARTERY" | AIF.shift == "VEIN")
     aif.shifted <- aif.shift.func(map.times, aif,
     param.est.whole.roi.xTofts$tlag)
     if (show.rt.fits == TRUE) {
     if (AIF.shift == "ARTERY") {
     dev.new(width = 6, height = 6, xpos = 1500,
     ypos = 0)
     plot(map.times, aif, ylim = c(0, max(aif,
     aif.shifted)), xlab = "min", ylab = "contrast agent",
     main = paste("Shifted Arterial Input Function (",
     format(param.est.whole.roi.xTofts$tlag *
     60, digits = 3), " sec)", sep = ""),
     type = "n")
     lines(map.times, aif, col = "red")
     lines(map.times, aif.shifted)
     legend(x = 0.55 * max(map.times), y = max(aif,
     aif.shifted), c("raw AIF", "shifted AIF"),
     c("red", "black"))
     }
     if (AIF.shift == "VEIN") {
     dev.new(width = 6, height = 6, xpos = 1500,
     ypos = 0)
     plot(map.times, aif, ylim = c(0, max(aif,
     aif.shifted)), xlab = "min", ylab = "contrast agent",
     main = paste("Shifted Venous Input Function (",
     format(param.est.whole.roi.xTofts$tlag *
     60, digits = 3), " sec)", sep = ""),
     type = "n")
     lines(map.times, aif, col = "blue")
     lines(map.times, aif.shifted)
     legend(x = 0.55 * max(map.times), y = max(aif,
     aif.shifted), c("raw VIF", "shifted VIF"),
     c("blue", "black"))
     }
     if (AIF.shift == "NONE") {
     dev.new(width = 6, height = 6, xpos = 1500,
     ypos = 0)
     plot(map.times, aif, ylim = c(0, max(aif)),
     xlab = "min", ylab = "contrast agent",
     main = "Vascular Input Function", type = "n")
     lines(map.times, aif)
     legend(x = 0.55 * max(map.times), y = max(aif),
     c("raw VIF"), c("black"))
     }
     }
     if (show.rt.fits == TRUE) {
     dev.new(width = 6, height = 6, xpos = 1500,
     ypos = 575)
     plot(map.times, roi, xlab = "min", ylab = "contrast agent",
     main = "median contrast agent conc; Tofts=blue, xTofts=red",
     cex = 2)
     text(x = 0.65 * max(map.times), y = 0.4 * max(roi,
     na.rm = TRUE), labels = "EXTENDED TOFTS PARAMS",
     cex = 1)
     if (class(fit.roi.median.xTofts) != "try-error") {
     if (fit.roi.median.xTofts$convergence ==
     0) {
     text(x = 0.65 * max(map.times), y = 0.35 *
     max(roi, na.rm = TRUE), labels = paste(expression(K^trans),
     " = ", format(param.est.whole.roi.xTofts$Ktrans,
     digits = 3), " 1/min (", cv.whole.roi.xTofts[1],
     "%)", sep = ""), cex = 1)
     text(x = 0.65 * max(map.times), y = 0.3 *
     max(roi, na.rm = TRUE), labels = paste(expression(k_ep),
     " = ", format(param.est.whole.roi.xTofts$kep,
     digits = 3), " 1/min (", cv.whole.roi.xTofts[2],
     "%)", sep = ""), cex = 1)
     text(x = 0.65 * max(map.times), y = 0.25 *
     max(roi, na.rm = TRUE), labels = paste(expression(v_b),
     " = ", format(param.est.whole.roi.xTofts$vb,
     digits = 3), " dimensionless (", cv.whole.roi.xTofts[3],
     "%)", sep = ""), cex = 1)
     if (AIF.shift != "NONE")
     text(x = 0.65 * max(map.times), y = 0.2 *
     max(roi, na.rm = TRUE), labels = paste(expression(t_lag),
     " = ", format(60 * param.est.whole.roi.xTofts$tlag,
     digits = 3), " sec (", cv.whole.roi.xTofts[4],
     "%)", sep = ""), cex = 1)
     lines(map.times, roi.median.fitted.xTofts,
     col = "red", lwd = 5)
     }
     }
     if (class(fit.roi.median.Tofts) != "try-error") {
     if (fit.roi.median.Tofts$convergence == 0) {
     lines(map.times, roi.median.fitted.Tofts,
     col = "blue", lwd = 2)
     }
     }
     }
     }
     if (file.original == FALSE) {
     fix.tlag <- TRUE
     if (file.format == "matlab") {
     mat_data <- readMat(file)
     args <- mat_data$args[, , 1]
     results_file <- args$resultsfile
     ID.visit <- args$IDvisit
     slice <- args$slice
     method.optimization <- args$methodoptimization
     show.rt.fits <- args$showrtfits
     param.for.avdt <- param.for.avdt
     range.map <- args$rangemap
     cutoff.map <- args$cutoffmap
     lo <- args$lo
     est.per.voxel.tlag <- args$estpervoxeltlag
     p0.xT <- mat_data$p0xT
     p0.T <- mat_data$p0T
     AIF.shift <- args$AIFshift
     nx <- mat_data$nx
     ny <- mat_data$ny
     nt <- mat_data$nt
     map_cc_slice <- mat_data$cc
     map_cc_roi <- mat_data$ccroi
     map.times <- mat_data$maptimes
     aif <- mat_data$aif
     aif.shifted <- mat_data$aifshifted
     map.KtransxT <- mat_data$mapKtransxT
     map.tlagxT <- mat_data$maptlagxT
     map.kepxT <- mat_data$mapkepxT
     map.vexT <- mat_data$mapvexT
     map.vbxT <- mat_data$mapvbxT
     map.KtransT.cv <- mat_data$mapKtransTcv
     map.kepT.cv <- mat_data$mapkepTcv
     map.KtransxT.cv <- mat_data$mapKtransxTcv
     map.tlagxT.cv <- mat_data$maptlagxTcv
     map.kepxT.cv <- mat_data$mapkepxTcv
     map.vbxT.cv <- mat_data$mapvbxTcv
     map.fitfailuresxT <- mat_data$mapfitfailuresxT
     map.KtransT <- mat_data$mapKtransT
     map.kepT <- mat_data$mapkepT
     map.veT <- mat_data$mapveT
     map.fitfailuresT <- mat_data$mapfitfailuresT
     mask.roi <- mat_data$maskroi
     param.est.medianT <- mat_data$paramestmedianT
     param.est.medianxT <- mat_data$paramestmedianxT
     cc.median <- mat_data$ccmedian
     roi.median.fitted <- mat_data$roimedianfitted
     param.est.whole.roi <- mat_data$paramestwholeroi
     map.AIC.compare <- mat_data$mapAICcompare
     map.AIC.T <- mat_data$mapAICT
     map.EF <- mat_data$mapEF
     map.AIC.xT <- mat_data$mapAICxT
     roi.median.fitted.Tofts <- mat_data$roimedianfittedTofts
     param.est.whole.roi.Tofts <- mat_data$paramestwholeroiTofts
     roi.median.fitted.xTofts <- mat_data$roimedianfittedxTofts
     param.est.whole.roi.xTofts <- mat_data$paramestwholeroixTofts
     cv.whole.roi.xTofts <- mat_data$cvwholeroixTofts
     params.xTofts <- mat_data$paramsxTofts
     rm(mat_data)
     }
     if (file.format == "RData") {
     load(file)
     args <- dcemri.data$args
     results_file <- args$resultsfile
     ID.visit <- args$IDvisit
     slice <- args$slice
     method.optimization <- args$methodoptimization
     show.rt.fits <- args$showrtfits
     param.for.avdt <- param.for.avdt
     range.map <- args$rangemap
     cutoff.map <- args$cutoffmap
     lo <- args$lo
     p0.xT <- dcemri.data$p0xT
     p0.T <- dcemri.data$p0T
     AIF.shift <- args$AIFshift
     est.per.voxel.tlag <- args$estpervoxeltlag
     nx <- dcemri.data$nx
     ny <- dcemri.data$ny
     nt <- dcemri.data$nt
     export.matlab <- args$exportmatlab
     map_cc_slice <- dcemri.data$cc
     map_cc_roi <- dcemri.data$ccroi
     map.times <- dcemri.data$maptimes
     aif <- dcemri.data$aif
     aif.shifted <- dcemri.data$aifshifted
     map.KtransT.cv <- dcemri.data$mapKtransTcv
     map.kepT.cv <- dcemri.data$mapkepTcv
     map.KtransxT.cv <- dcemri.data$mapKtransxTcv
     map.tlagxT.cv <- dcemri.data$maptlagxTcv
     map.kepxT.cv <- dcemri.data$mapkepxTcv
     map.vbxT.cv <- dcemri.data$mapvbxTcv
     map.KtransxT <- dcemri.data$mapKtransxT
     map.tlagxT <- dcemri.data$maptlagxT
     map.kepxT <- dcemri.data$mapkepxT
     map.vexT <- dcemri.data$mapvexT
     map.vbxT <- dcemri.data$mapvbxT
     map.fitfailuresxT <- dcemri.data$mapfitfailuresxT
     map.KtransT <- dcemri.data$mapKtransT
     map.kepT <- dcemri.data$mapkepT
     map.veT <- dcemri.data$mapveT
     map.fitfailuresT <- dcemri.data$mapfitfailuresT
     mask.roi <- dcemri.data$maskroi
     param.est.medianT <- dcemri.data$paramestmedianT
     param.est.medianxT <- dcemri.data$paramestmedianxT
     cc.median <- dcemri.data$ccmedian
     roi.median.fitted <- dcemri.data$roimedianfitted
     param.est.whole.roi <- dcemri.data$paramestwholeroi
     map.AIC.compare <- dcemri.data$mapAICcompare
     map.AIC.T <- dcemri.data$mapAICT
     map.EF <- dcemri.data$mapEF
     map.AIC.xT <- dcemri.data$mapAICxT
     roi.median.fitted.Tofts <- dcemri.data$roimedianfittedTofts
     param.est.whole.roi.Tofts <- dcemri.data$paramestwholeroiTofts
     roi.median.fitted.xTofts <- dcemri.data$roimedianfittedxTofts
     param.est.whole.roi.xTofts <- dcemri.data$paramestwholeroixTofts
     cv.whole.roi.xTofts <- dcemri.data$cvwholeroixTofts
     params.xTofts <- dcemri.data$paramsxTofts
     rm(dcemri.data)
     }
     modeltype1 <- "xTofts"
     modeltype2 <- "Tofts"
     }
     if (file.original == TRUE) {
     cat("fitting", modeltype1, "and", modeltype2,
     "models to ROI voxels...", "\n")
     ptm <- proc.time()[3]
     t <- map.times
     map.KtransxT <- matrix(NA, nrow = nx, ncol = ny)
     map.kepxT <- matrix(NA, nrow = nx, ncol = ny)
     map.vexT <- matrix(NA, nrow = nx, ncol = ny)
     map.vbxT <- matrix(NA, nrow = nx, ncol = ny)
     map.tlagxT <- matrix(NA, nrow = nx, ncol = ny)
     map.KtransxT.cv <- matrix(NA, nrow = nx, ncol = ny)
     map.kepxT.cv <- matrix(NA, nrow = nx, ncol = ny)
     map.vbxT.cv <- matrix(NA, nrow = nx, ncol = ny)
     map.tlagxT.cv <- matrix(NA, nrow = nx, ncol = ny)
     map.fitfailuresxT <- matrix(NA, nrow = nx, ncol = ny)
     map.OptimValuexT <- matrix(NA, nrow = nx, ncol = ny)
     map.KtransT <- matrix(NA, nrow = nx, ncol = ny)
     map.kepT <- matrix(NA, nrow = nx, ncol = ny)
     map.veT <- matrix(NA, nrow = nx, ncol = ny)
     map.KtransT.cv <- matrix(NA, nrow = nx, ncol = ny)
     map.kepT.cv <- matrix(NA, nrow = nx, ncol = ny)
     map.fitfailuresT <- matrix(NA, nrow = nx, ncol = ny)
     map.OptimValueT <- matrix(NA, nrow = nx, ncol = ny)
     map.AIC.compare <- matrix(0, nrow = nx, ncol = ny)
     map.AIC.T <- matrix(NA, nrow = nx, ncol = ny)
     map.EF <- matrix(NA, nrow = nx, ncol = ny)
     map.AIC.xT <- matrix(NA, nrow = nx, ncol = ny)
     cc_fittedxT <- array(0, dim = c(nx, ny, nt))
     cc_fittedT <- array(0, dim = c(nx, ny, nt))
     nv <- 1
     nv1_q <- trunc(quantile(1:length(mask.roi[mask.roi ==
     1]), probs = c(0.1, 0.2, 0.3, 0.4, 0.5, 0.6,
     0.7, 0.8, 0.9, 1)))
     if (show.rt.fits == TRUE)
     dev.new(xpos = 3500, ypos = 0)
     ptm_slice <- proc.time()[3]
     GTzero <- function(x) {
     length(as.vector(x > 0)[as.vector(x > 0) ==
     TRUE])/length(x)
     }
     for (x in 1:nx) {
     for (y in 1:ny) {
     if (mask.roi[x, y] == 1 & GTzero(map_cc_slice[x,
     y, ]) <= fracGTzero) {
     map.fitfailuresxT[x, y] <- -2
     map.fitfailuresT[x, y] <- -2
     }
     if (mask.roi[x, y] == 1 & GTzero(map_cc_slice[x,
     y, ]) > fracGTzero)
     map.EF[x, y] <- 1
     if (mask.roi[x, y] == 1 & GTzero(map_cc_slice[x,
     y, ]) > fracGTzero) {
     nv <- nv + 1
     roi <- map_cc_slice[x, y, ]
     fix.tlag <- TRUE
     roi.model <- roi.modelxT
     if (verbose == TRUE) {
     cat("x =", x, "\n")
     cat("y =", y, "\n")
     cat("contrast agent curve =", roi, "\n")
     cat("fitting xTofts model to voxel data...")
     }
     if (method.optimization == "L-BFGS-B")
     fit_roi <- try(optim(p0.xT, Obj_roi,
     model = roi.model, t = map.times, dt = diff(map.times),
     cp = aif, roi = roi, method = "L-BFGS-B",
     lower = c(lo, lo, lo), upper = c(Inf,
     Inf, Inf), hessian = TRUE), silent = try.silent)
     if (method.optimization != "L-BFGS-B")
     fit_roi <- try(optim(p0.xT, Obj_roi,
     model = roi.model, t = map.times, dt = diff(map.times),
     cp = aif, roi = roi, method = method.optimization,
     hessian = TRUE), silent = try.silent)
     if (verbose == TRUE)
     cat("done", "\n")
     if (class(fit_roi) != "try-error") {
     if (fit_roi$convergence == 0) {
     RSS <- sum((roi.model(p = fit_roi$par,
     t = map.times, dt = diff(map.times),
     cp = aif) - roi)^2)
     param_est <- fit_roi$par
     nD <- nt
     nP <- length(param_est)
     hess <- fit_roi$hessian
     cov <- try((nP * RSS/(nD - nP)) * solve(hess),
     silent = try.silent)
     if (class(cov) != "try-error") {
     sd <- try(sqrt(diag(cov)), silent = try.silent)
     if (class(sd) != "try-error") {
     cv <- sd/param_est * 100
     cv.roi.voxel <- as.numeric(format(cv,
     digits = 1))
     map.KtransxT.cv[x, y] <- cv.roi.voxel[1]
     map.kepxT.cv[x, y] <- cv.roi.voxel[2]
     map.vbxT.cv[x, y] <- cv.roi.voxel[3]
     if (est.per.voxel.tlag == TRUE)
     map.tlagxT.cv[x, y] <- cv.roi.voxel[4]
     }
     }
     map.KtransxT[x, y] <- fit_roi$par[1]
     map.kepxT[x, y] <- fit_roi$par[2]
     if (map.kepxT[x, y] < 1e-05)
     map.kepxT[x, y] <- 1e-05
     map.vexT[x, y] <- fit_roi$par[1]/map.kepxT[x,
     y]
     map.vbxT[x, y] <- fit_roi$par[3]
     if (est.per.voxel.tlag == TRUE)
     map.tlagxT[x, y] <- fit_roi$par[4]
     map.OptimValuexT[x, y] <- fit_roi$value
     }
     }
     if (class(fit_roi) == "try-error") {
     if (verbose == TRUE)
     cat("...but a <try-error> occurred",
     "\n")
     map.fitfailuresxT[x, y] <- 99
     }
     if (class(fit_roi) != "try-error")
     map.fitfailuresxT[x, y] <- fit_roi$convergence
     if (class(fit_roi) == "try-error") {
     map.KtransxT[x, y] <- NA
     map.kepxT[x, y] <- NA
     map.vexT[x, y] <- NA
     map.vbxT[x, y] <- NA
     map.tlagxT[x, y] <- NA
     }
     if (class(fit_roi) != "try-error") {
     if (fit_roi$convergence > 0) {
     map.KtransxT[x, y] <- NA
     map.kepxT[x, y] <- NA
     map.vexT[x, y] <- NA
     map.vbxT[x, y] <- NA
     map.tlagxT[x, y] <- NA
     }
     }
     if (verbose == TRUE)
     cat("simulating xTofts model at estimated parameter values...")
     if (class(fit_roi) != "try-error") {
     if (fit_roi$convergence == 0) {
     simulation <- roi.model(p = fit_roi$par,
     t = map.times, dt = diff(map.times),
     cp = aif)
     cc_fittedxT[x, y, ] <- simulation
     }
     }
     if (verbose == TRUE)
     cat("done", "\n")
     roi.model <- roi.modelT
     if (verbose == TRUE)
     cat("fitting Tofts model to voxel data...")
     if (method.optimization == "L-BFGS-B")
     fit_roiT <- try(optim(p0.T, Obj_roi,
     model = roi.model, t = map.times, dt = diff(map.times),
     cp = aif, roi = roi, method = "L-BFGS-B",
     lower = lower.wholeT, upper = upper.wholeT,
     hessian = TRUE), silent = try.silent)
     if (method.optimization != "L-BFGS-B")
     fit_roiT <- try(optim(p0.T, Obj_roi,
     model = roi.model, t = map.times, dt = diff(map.times),
     cp = aif, roi = roi, method = method.optimization,
     hessian = TRUE), silent = try.silent)
     if (verbose == TRUE)
     cat("done", "\n")
     if (class(fit_roiT) != "try-error") {
     if (fit_roiT$convergence == 0) {
     RSS <- sum((roi.model(p = fit_roiT$par,
     t = map.times, dt = diff(map.times),
     cp = aif) - roi)^2)
     nD <- nt
     nP <- length(param_est)
     param_est <- fit_roiT$par
     df <- nt - length(param_est)
     hess <- fit_roiT$hessian
     cov <- try((nP * RSS/(nD - nP)) * solve(hess),
     silent = try.silent)
     if (class(cov) != "try-error") {
     sd <- try(sqrt(diag(cov)), silent = try.silent)
     if (class(sd) != "try-error") {
     cv <- sd/param_est * 100
     cv.roi.voxel <- as.numeric(format(cv,
     digits = 1))
     map.KtransT.cv[x, y] <- cv.roi.voxel[1]
     map.kepT.cv[x, y] <- cv.roi.voxel[2]
     }
     }
     map.KtransT[x, y] <- fit_roiT$par[1]
     map.kepT[x, y] <- fit_roiT$par[2]
     if (map.kepT[x, y] < 1e-05)
     map.kepT[x, y] <- 1e-05
     map.veT[x, y] <- fit_roiT$par[1]/map.kepT[x,
     y]
     map.OptimValueT[x, y] <- fit_roiT$value
     }
     }
     if (class(fit_roiT) == "try-error") {
     map.fitfailuresT[x, y] <- 99
     if (verbose == TRUE)
     cat("...but a <try-error> occurred",
     "\n")
     }
     if (class(fit_roiT) == "try-error") {
     map.KtransT[x, y] <- NA
     map.kepT[x, y] <- NA
     map.veT[x, y] <- NA
     }
     if (class(fit_roiT) != "try-error") {
     if (fit_roiT$convergence > 0) {
     map.KtransT[x, y] <- NA
     map.kepT[x, y] <- NA
     map.veT[x, y] <- NA
     }
     }
     if (class(fit_roiT) != "try-error")
     map.fitfailuresT[x, y] <- fit_roiT$convergence
     if (verbose == TRUE)
     cat("simulating Tofts model at estimated parameter values...")
     if (class(fit_roiT) != "try-error") {
     if (fit_roiT$convergence == 0) {
     simulation_2 <- roi.model(p = fit_roiT$par,
     t = map.times, dt = diff(map.times),
     cp = aif)
     cc_fittedT[x, y, ] <- simulation_2
     }
     }
     if (verbose == TRUE)
     cat("done", "\n")
     roi.model <- roi.modelxT
     if (class(fit_roi) != "try-error" & class(fit_roiT) !=
     "try-error") {
     if (fit_roi$convergence == 0 & fit_roiT$convergence ==
     0) {
     if (verbose == TRUE)
     cat("calculating AICc values...")
     simulation_AIC <- simulation
     simulation_AIC_2 <- simulation_2
     np_1 <- length(fit_roiT$par)
     np_2 <- length(fit_roi$par) + 1
     AIC1 <- nt * log(fit_roiT$value) +
     2 * (np_1 + 1) + (2 * (np_1 + 1) *
     (np_1 + 2))/(nt - np_1 - 2)
     AIC2 <- nt * log(fit_roi$value) + 2 *
     (np_2 + 1) + (2 * (np_2 + 1) * (np_2 +
     2))/(nt - np_2 - 2)
     AIC1 <- as.numeric(format(AIC1, digits = 1))
     AIC2 <- as.numeric(format(AIC2, digits = 1))
     if (AIC2 < AIC1)
     map.AIC.compare[x, y] <- 1
     if (AIC2 >= AIC1)
     map.AIC.compare[x, y] <- 2
     map.AIC.T[x, y] <- AIC1
     map.AIC.xT[x, y] <- AIC2
     }
     }
     if (class(fit_roi) != "try-error") {
     if (fit_roi$convergence > 0) {
     map.AIC.compare[x, y] <- NA
     }
     }
     if (class(fit_roiT) != "try-error") {
     if (fit_roiT$convergence > 0) {
     map.AIC.compare[x, y] <- NA
     }
     }
     if (class(fit_roi) == "try-error" | class(fit_roiT) ==
     "try-error")
     map.AIC.compare[x, y] <- NA
     if (verbose == TRUE) {
     cat("done", "\n")
     cat("======================", "\n")
     }
     if (show.rt.fits == TRUE) {
     if (class(fit_roiT) != "try-error" &
     class(fit_roi) != "try-error") {
     if (fit_roiT$convergence == 0 & fit_roi$convergence ==
     0) {
     plot(map.times, roi, ylab = "contrast agent",
     xlab = "min", main = paste(modeltype1,
     "(red) and", modeltype2, "(blue)"),
     cex = 3)
     lines(map.times, simulation, col = "red",
     lwd = 5)
     lines(map.times, simulation_2, col = "blue",
     lwd = 2)
     }
     }
     }
     if (nv == 2)
     cat("progress: ")
     if (nv == nv1_q[[1]])
     cat("10%..")
     if (nv == nv1_q[[2]])
     cat("20%..")
     if (nv == nv1_q[[3]])
     cat("30%..")
     if (nv == nv1_q[[4]])
     cat("40%..")
     if (nv == nv1_q[[5]])
     cat("50%..")
     if (nv == nv1_q[[6]])
     cat("60%..")
     if (nv == nv1_q[[7]])
     cat("70%..")
     if (nv == nv1_q[[8]])
     cat("80%..")
     if (nv == nv1_q[[9]])
     cat("90%..", "\n")
     if (nv == nv1_q[[10]] - 10)
     cat("..10")
     if (nv == nv1_q[[10]] - 9)
     cat("..9..")
     if (nv == nv1_q[[10]] - 8)
     cat("8..")
     if (nv == nv1_q[[10]] - 7)
     cat("7..")
     if (nv == nv1_q[[10]] - 6)
     cat("6..")
     if (nv == nv1_q[[10]] - 5)
     cat("5..")
     if (nv == nv1_q[[10]] - 4)
     cat("4..")
     if (nv == nv1_q[[10]] - 3)
     cat("3..")
     if (nv == nv1_q[[10]] - 2)
     cat("2..")
     if (nv == nv1_q[[10]] - 1)
     cat("1..", "\n")
     }
     }
     }
     cat("..done in", format((proc.time()[3] - ptm)/60,
     digits = 2), "minutes.", "\n")
     cat("--------", "\n")
     graphics.off()
     KtransxT.median <- median(map.KtransxT[map.KtransxT >
     0 & map.fitfailuresxT == 0], na.rm = TRUE)
     kepxT.median <- median(map.kepxT[map.kepxT >
     0 & map.fitfailuresxT == 0], na.rm = TRUE)
     vexT.median <- median(map.vexT[map.vexT > 0 &
     map.fitfailuresxT == 0], na.rm = TRUE)
     vbxT.median <- median(map.vbxT[map.vbxT >= 0 &
     map.fitfailuresxT == 0], na.rm = TRUE)
     if (est.per.voxel.tlag == TRUE)
     tlagxT.median <- median(map.tlagxT[map.tlagxT >=
     0 & map.fitfailuresxT == 0], na.rm = TRUE)
     if (est.per.voxel.tlag == FALSE)
     tlagxT.median <- NA
     fitfailuresxT.total <- length(map.fitfailuresxT[map.fitfailuresxT >=
     1])/length(map.fitfailuresxT[map.fitfailuresxT >=
     0]) * 100
     param.est.medianxT <- list(KtransxT.median, kepxT.median,
     vexT.median, vbxT.median, tlagxT.median, fitfailuresxT.total)
     names(param.est.medianxT) <- c("Ktrans.median",
     "kep.median", "ve.median", "vb.median", "tlag.median",
     "percent.fitfailures")
     KtransT.median <- median(map.KtransT[map.KtransT >
     0 & map.fitfailuresT == 0], na.rm = TRUE)
     kepT.median <- median(map.kepT[map.kepT > 0 &
     map.fitfailuresT == 0], na.rm = TRUE)
     veT.median <- median(map.veT[map.veT > 0 & map.fitfailuresT ==
     0], na.rm = TRUE)
     fitfailuresT.total <- length(map.fitfailuresT[map.fitfailuresT >=
     1])/length(map.fitfailuresT[map.fitfailuresT >=
     0]) * 100
     param.est.medianT <- list(KtransT.median, kepT.median,
     veT.median, fitfailuresT.total)
     names(param.est.medianT) <- c("Ktrans.median",
     "kep.median", "ve.median", "percent.fitfailures")
     }
     if (file.original == TRUE) {
     if (file == "concatenate.KAT.with.KAT.checkData.RData")
     ID.visit <- paste(strsplit(results_file, split = ".mat")[[1]],
     "_s", slice, sep = "")
     if (file != "concatenate.KAT.with.KAT.checkData.RData")
     ID.visit <- paste(strsplit(file, split = ".mat")[[1]],
     "_s", slice, sep = "")
     }
     if (file.original == FALSE)
     ID.visit <- strsplit(file, split = ".mat")[[1]]
     ID.visit <- strsplit(ID.visit, split = "/")
     ID.visit <- ID.visit[[1]][length(ID.visit[[1]])]
     IDvp <- strsplit(ID.visit, split = "_")
     ID.visit.forplot <- paste(IDvp[[1]][1], ".", IDvp[[1]][2],
     ".", IDvp[[1]][3], ".", IDvp[[1]][4], sep = "")
     DATE <- date()
     if (length(strsplit(DATE, split = " ")[[1]]) ==
     2) {
     full_date <- strsplit(DATE, split = " ")[[1]]
     full_date_1 <- full_date[1]
     full_date_1 <- strsplit(full_date_1, split = " ")[[1]]
     full_date_2 <- full_date[2]
     full_date_2 <- strsplit(full_date_2, split = " ")[[1]]
     month <- full_date_1[2]
     day <- full_date_2[1]
     time <- full_date_2[2]
     year <- full_date_2[3]
     }
     if (length(strsplit(DATE, split = " ")[[1]]) ==
     1) {
     full_date <- strsplit(DATE, split = " ")[[1]]
     month <- full_date[2]
     day <- full_date[3]
     time <- full_date[4]
     year <- full_date[5]
     }
     year <- strsplit(year, split = "")[[1]]
     year <- paste(year[3], year[4], sep = "")
     time_concat <- strsplit(time, split = ":")[[1]]
     time_concat <- paste(paste(time_concat[1], time_concat[2],
     sep = ""), time_concat[3], sep = "")
     DATE <- paste(paste(paste(paste(day, month, sep = ""),
     year, sep = ""), "-", sep = ""), time_concat,
     sep = "")
     filename3 <- paste(ID.visit, "_KAT_", DATE, ".mat",
     sep = "")
     filename3 <- sub(".RData", "", filename3)
     if (file.original == FALSE) {
     roi.model <- roi.modelxT
     }
     if (param.for.avdt == "Ktrans")
     MAP <- map.KtransxT
     if (param.for.avdt == "kep")
     MAP <- map.kepxT
     if (param.for.avdt == "ve")
     MAP <- map.vexT
     if (param.for.avdt == "vb")
     MAP <- map.vbxT
     if (param.for.avdt == "tlag")
     MAP <- map.tlagxT
     x_min <- 0
     x_max <- nx
     y_min <- 0
     y_max <- ny
     for (xx in 1:nx) {
     if (sum(MAP[xx, ], na.rm = TRUE) > 0) {
     x_min <- xx - 3
     break
     }
     if (sum(MAP[xx, ], na.rm = TRUE) == 0)
     xx <- xx + 1
     }
     for (y in 1:ny) {
     if (sum(MAP[, y], na.rm = TRUE) > 0) {
     y_min <- y - 3
     break
     }
     if (sum(MAP[, y], na.rm = TRUE) == 0)
     y <- y + 1
     }
     for (xx in nx:0) {
     if (sum(MAP[xx, ], na.rm = TRUE) > 0) {
     x_max <- xx + 3
     break
     }
     if (sum(MAP[xx, ], na.rm = TRUE) == 0)
     xx <- xx - 1
     }
     for (y in ny:0) {
     if (sum(MAP[, y], na.rm = TRUE) > 0) {
     y_max <- y + 3
     break
     }
     if (sum(MAP[, y], na.rm = TRUE) == 0)
     y <- y - 1
     }
     MAP_ul_s1 <- MAP[MAP > 0]
     MAP_ul_s1 <- sort(MAP_ul_s1)
     MAP_ul <- range.map * (max(MAP_ul_s1[1:length(MAP_ul_s1) *
     cutoff.map]))
     if (file.original == FALSE) {
     if (param.for.avdt == "Ktrans")
     MAP <- map.KtransxT
     if (param.for.avdt == "kep")
     MAP <- map.kepxT
     if (param.for.avdt == "ve")
     MAP <- map.vexT
     if (param.for.avdt == "vb")
     MAP <- map.vbxT
     if (param.for.avdt == "tlag")
     MAP <- map.tlagxT
     if (file.format == "matlab") {
     if (param.for.avdt == "Ktrans")
     param.median <- param.est.medianxT[, , 1]$Ktrans.median[1]
     if (param.for.avdt == "kep")
     param.median <- param.est.medianxT[, , 1]$kep.median[1]
     if (param.for.avdt == "ve")
     param.median <- param.est.medianxT[, , 1]$ve.median[1]
     if (param.for.avdt == "vb")
     param.median <- param.est.medianxT[, , 1]$vb.median[1]
     if (param.for.avdt == "tlag")
     param.median <- param.est.medianxT[, , 1]$tlag.median[1]
     }
     if (file.format == "RData") {
     if (param.for.avdt == "Ktrans")
     param.median <- param.est.medianxT$Ktrans.median
     if (param.for.avdt == "kep")
     param.median <- param.est.medianxT$kep.median
     if (param.for.avdt == "ve")
     param.median <- param.est.medianxT$ve.median
     if (param.for.avdt == "vb")
     param.median <- param.est.medianxT$vb.median
     if (param.for.avdt == "tlag")
     param.median <- param.est.medianxT$tlag.median
     }
     MAP_for_plot <- MAP
     MAP_for_plot[MAP_for_plot < 0] <- 0
     MAP_for_plot[MAP_for_plot >= MAP_ul * 0.99] <- MAP_ul *
     0.99
     dev.new(width = 6, height = 6, xpos = 1860, ypos = 578)
     plot(map.times, cc.median, xlab = "min", ylab = "contrast agent",
     main = paste(modeltype1, "(red) and", modeltype2,
     "(blue) fitted to median whole ROI data"),
     cex.main = 1, cex.axis = 1, cex.lab = 1, cex = 2)
     lines(map.times, roi.median.fitted.xTofts, col = "red",
     lwd = 5)
     lines(map.times, roi.median.fitted.Tofts, col = "blue",
     lwd = 2)
     text(x = 0.6 * max(map.times), y = 0.3 * max(cc.median,
     na.rm = TRUE), labels = "EXTENDED TOFTS PARAMS",
     cex = 1)
     text(x = 0.6 * max(map.times), y = 0.24 * max(cc.median,
     na.rm = TRUE), labels = paste(expression(K^trans),
     " = ", format(param.est.whole.roi.xTofts$Ktrans,
     digits = 3), " 1/min (", cv.whole.roi.xTofts[1],
     "%)", sep = ""), cex = 1)
     text(x = 0.6 * max(map.times), y = 0.18 * max(cc.median,
     na.rm = TRUE), labels = paste(expression(k_ep),
     " = ", format(param.est.whole.roi.xTofts$kep,
     digits = 3), " 1/min (", cv.whole.roi.xTofts[2],
     "%)", sep = ""), cex = 1)
     text(x = 0.6 * max(map.times), y = 0.12 * max(cc.median,
     na.rm = TRUE), labels = paste(expression(v_b),
     " = ", format(param.est.whole.roi.xTofts$vb,
     digits = 3), " dimensionless (", cv.whole.roi.xTofts[3],
     "%)", sep = ""), cex = 1)
     if (AIF.shift != "NONE")
     text(x = 0.6 * max(map.times), y = 0.06 * max(cc.median,
     na.rm = TRUE), labels = paste(expression(t_lag),
     " = ", format(60 * param.est.whole.roi.xTofts$tlag,
     digits = 3), " sec (", cv.whole.roi.xTofts[4],
     "%)", sep = ""), cex = 1)
     dev.new(width = 6, height = 6, xpos = 1860, ypos = 0)
     image(map.AIC.compare, xlim = c(x_min/nx, x_max/nx),
     ylim = c(y_min/ny, y_max/ny), col = palette(colorRampPalette(c("black",
     "red", "blue"))(3)), main = paste("best fit:",
     modeltype1, "(red) and", modeltype2, "(blue)"))
     image(map.AIC.compare, xlim = c(x_min/nx, x_max/nx),
     ylim = c(y_min/ny, y_max/ny), col = palette(colorRampPalette(c("black",
     "red", "blue"))(3)), main = paste("best fit:",
     modeltype1, "(red) and", modeltype2, "(blue)"))
     dev.new(width = 12.75, height = 12.75, xpos = 238,
     ypos = 0)
     image(MAP_for_plot, col = palette(colorRampPalette(c("black",
     "red", "yellow"))(1000)), xlim = c(x_min/nx,
     x_max/nx), ylim = c(y_min/ny, y_max/ny), zlim = c(0,
     MAP_ul), main = paste("Model Type=", modeltype1,
     " Median ", param.for.avdt, "=", format(param.median,
     digit = 2), " VisitID=", strsplit(args$IDvisit,
     split = "_")[[1]][1], " slice=", args$slice,
     sep = ""))
     image(MAP_for_plot, col = palette(colorRampPalette(c("black",
     "red", "yellow"))(1000)), xlim = c(x_min/nx,
     x_max/nx), ylim = c(y_min/ny, y_max/ny), zlim = c(0,
     MAP_ul), main = paste("Model Type=", modeltype1,
     " Median ", param.for.avdt, "=", format(param.median,
     digit = 2), " VisitID=", strsplit(args$IDvisit,
     split = "_")[[1]][1], " slice=", args$slice,
     sep = ""))
     text(x_min/nx + (x_max/nx - x_min/nx) * 0.97,
     y_min/ny + (y_max/ny - y_min/ny) * 0.99, "close",
     col = "green")
     text(x_min/nx + (x_max/nx - x_min/nx) * 0.05,
     y_min/ny + (y_max/ny - y_min/ny) * 0.99, "print to PDF",
     col = "green")
     text(x_min/nx + (x_max/nx - x_min/nx) * 0.84,
     y_min/ny + (y_max/ny - y_min/ny) * 0.01, paste("KAT for DCEMRI v",
     KAT.version, ", Genentech PTPK", sep = ""),
     col = "darkgrey")
     legend <- seq(0, MAP_ul, by = 0.001)
     dim(legend) <- c(1, length(legend))
     dev.new(width = 2.5, height = 12.75, xpos = 0,
     ypos = 0)
     image(legend, col = palette(colorRampPalette(c("black",
     "red", "yellow"))(1000)), zlim = c(0, MAP_ul),
     xaxt = "n", yaxt = "n", xlab = "0", main = format(MAP_ul,
     digits = 3), ylab = expression(min^-1), cex.lab = 1.25,
     cex.main = 1.05)
     image(legend, col = palette(colorRampPalette(c("black",
     "red", "yellow"))(1000)), zlim = c(0, MAP_ul),
     xaxt = "n", yaxt = "n", xlab = "0", main = format(MAP_ul,
     digits = 3), ylab = expression(min^-1), cex.lab = 1.25,
     cex.main = 1.05)
     if (AIF.shift == "ARTERY") {
     dev.new(width = 5.15, height = 4.4, xpos = 1500,
     ypos = 0)
     plot(map.times, aif, ylim = c(0, max(aif, aif.shifted)),
     xlab = "min", ylab = "contrast agent", main = paste("Shifted Arterial Input Function (",
     format(param.est.whole.roi.xTofts$tlag *
     60, digits = 3), " sec)", sep = ""),
     type = "n")
     lines(map.times, aif, col = "red")
     lines(map.times, aif.shifted)
     legend(x = 0.55 * max(map.times), y = max(aif,
     aif.shifted), c("raw AIF", "shifted AIF"),
     c("red", "black"))
     }
     if (AIF.shift == "VEIN") {
     dev.new(width = 5.15, height = 4.4, xpos = 1500,
     ypos = 0)
     plot(map.times, aif, ylim = c(0, max(aif, aif.shifted)),
     xlab = "min", ylab = "contrast agent", main = paste("Shifted Venous Input Function (",
     format(param.est.whole.roi.xTofts$tlag *
     60, digits = 3), " sec)", sep = ""),
     type = "n")
     lines(map.times, aif, col = "blue")
     lines(map.times, aif.shifted)
     legend(x = 0.55 * max(map.times), y = max(aif,
     aif.shifted), c("raw VIF", "shifted VIF"),
     c("blue", "black"))
     }
     if (AIF.shift == "NONE") {
     dev.new(width = 5.15, height = 4.4, xpos = 1500,
     ypos = 0)
     plot(map.times, aif, ylim = c(0, max(aif)),
     xlab = "min", ylab = "contrast agent", main = "Vascular Input Function",
     type = "n")
     lines(map.times, aif)
     legend(x = 0.55 * max(map.times), y = max(aif),
     c("raw VIF"), c("black"))
     }
     dev.set(4)
     inf <- 1
     newplot <- 1
     newplot2 <- 1
     legend_count <- 1
     legend_labels <- 1:1000
     legend_matrix <- matrix(0, ncol = ny, nrow = nx)
     cat("---", "\n")
     while (inf == 1) {
     z <- locator(1, type = "o", col = "green")
     xx <- round(z$x * (nx - 1) + 1)
     yy <- round(z$y * (ny - 1) + 1)
     if (legend_matrix[xx, yy] == 0) {
     legend(z$x, z$y, legend_labels[legend_count],
     col = "green", text.col = "green")
     legend_matrix[xx, yy] <- 1
     cat("Voxel Number/Coordinates: n=", legend_count,
     ", x=", xx, ", y=", yy, "\n", sep = "")
     cat("Parameter estimates (xTofts): Ktrans=",
     format(map.KtransxT[xx, yy], digits = 3),
     ", ve=", format(map.KtransxT[xx, yy]/map.kepxT[xx,
     yy], digits = 3), ", vb=", format(map.vbxT[xx,
     yy], digits = 3), ", tlag=", format(map.tlagxT[xx,
     yy], digits = 3), "\n", sep = "")
     cat("%CVs of Parameter estimates (xTofts): Ktrans=",
     format(map.KtransxT.cv[xx, yy], digits = 2),
     ", kep (Ktrans/ve)=", format(map.kepxT.cv[xx,
     yy], digits = 2), ", vb=", format(map.vbxT.cv[xx,
     yy], digits = 2), ", tlag=", format(map.tlagxT.cv[xx,
     yy], digits = 2), "\n", sep = "")
     cat("Parameter estimates (Tofts): Ktrans=",
     format(map.KtransT[xx, yy], digits = 3),
     ", ve=", format(map.KtransT[xx, yy]/map.kepT[xx,
     yy], digits = 3), "\n", sep = "")
     cat("%CVs of Parameter estimates (Tofts): Ktrans=",
     format(map.KtransT.cv[xx, yy], digits = 2),
     ", kep (Ktrans/ve)=", format(map.kepT.cv[xx,
     yy], digits = 3), "\n", sep = "")
     cat("---", "\n")
     legend_count <- legend_count + 1
     }
     xdim <- x_max - x_min
     ydim <- y_max - y_min
     if (xx > (x_max - 0.1 * xdim) && yy > (y_max -
     0.02 * ydim)) {
     graphics.off()
     dev.off()
     }
     xx_old <- xx
     yy_old <- yy
     conc <- 1:nt
     for (i in 1:nt) conc[i] <- map_cc_slice[xx,
     yy, i]
     if (xx < (x_min + 0.12 * xdim) && yy > (y_max -
     0.02 * ydim)) {
     pdf(file = paste(results_file, "-SUMMARY.pdf",
     sep = ""), height = 12, width = 15)
     layout(matrix(c(1, 2, 3, 4, 5, 6), 2, 3,
     byrow = TRUE), widths = c(1.5, 5.5, 5.5))
     par(omi = c(0.15, 0.15, 0.15, 0.15))
     image(legend, col = palette(colorRampPalette(c("black",
     "red", "yellow"))(1000)), zlim = c(0, MAP_ul),
     xaxt = "n", yaxt = "n", xlab = "0", main = format(MAP_ul,
     digits = 3), ylab = expression(min^-1),
     cex.lab = 1.5, cex.main = 1.5, cex.axis = 1.5)
     image(legend, col = palette(colorRampPalette(c("black",
     "red", "yellow"))(1000)), zlim = c(0, MAP_ul),
     xaxt = "n", yaxt = "n", xlab = "0", main = format(MAP_ul,
     digits = 3), ylab = expression(min^-1),
     cex.lab = 1.5, cex.main = 1.5, cex.axis = 1.5,
     add = TRUE)
     image(MAP_for_plot, col = palette(colorRampPalette(c("black",
     "red", "yellow"))(1000)), xlim = c(x_min/nx,
     x_max/nx), ylim = c(y_min/ny, y_max/ny),
     zlim = c(0, MAP_ul), main = paste("Model Type=",
     modeltype1, " Median ", param.for.avdt,
     "=", format(param.median, digit = 2),
     " VisitID=", strsplit(args$IDvisit,
     split = "_")[[1]][1], " slice=",
     args$slice, sep = ""), cex.axis = 1.5)
     image(MAP_for_plot, col = palette(colorRampPalette(c("black",
     "red", "yellow"))(1000)), xlim = c(x_min/nx,
     x_max/nx), ylim = c(y_min/ny, y_max/ny),
     zlim = c(0, MAP_ul), main = paste("Model Type=",
     modeltype1, " Median ", param.for.avdt,
     "=", format(param.median, digit = 2),
     " VisitID=", strsplit(args$IDvisit,
     split = "_")[[1]][1], " slice=",
     args$slice, sep = ""), cex.axis = 1.5,
     add = TRUE)
     plot(map.times, cc.median, xlab = "min",
     ylab = "contrast agent", main = "median contrast agent conc; Tofts=blue, xTofts=red",
     cex = 3, cex.axis = 1.5, cex.lab = 1.5)
     text(x = 0.6 * max(map.times), y = 0.25 *
     max(cc.median, na.rm = TRUE), labels = "EXTENDED TOFTS PARAMS",
     cex = 1.5)
     text(x = 0.6 * max(map.times), y = 0.2 *
     max(cc.median, na.rm = TRUE), labels = paste(expression(K^trans),
     " = ", format(param.est.whole.roi.xTofts$Ktrans,
     digits = 3), " 1/min (", cv.whole.roi.xTofts[1],
     "%)", sep = ""), cex = 1.5)
     text(x = 0.6 * max(map.times), y = 0.15 *
     max(cc.median, na.rm = TRUE), labels = paste(expression(v_e),
     " = ", format(param.est.whole.roi.xTofts$ve,
     digits = 3), " 1/min (", cv.whole.roi.xTofts[2],
     "%)", sep = ""), cex = 1.5)
     text(x = 0.6 * max(map.times), y = 0.1 *
     max(cc.median, na.rm = TRUE), labels = paste(expression(v_b),
     " = ", format(param.est.whole.roi.xTofts$vb,
     digits = 3), " dimensionless (", cv.whole.roi.xTofts[3],
     "%)", sep = ""), cex = 1.5)
     if (AIF.shift != "NONE")
     text(x = 0.6 * max(map.times), y = 0.05 *
     max(cc.median, na.rm = TRUE), labels = paste(expression(t_lag),
     " = ", format(60 * param.est.whole.roi.xTofts$tlag,
     digits = 3), " sec (", cv.whole.roi.xTofts[4],
     "%)", sep = ""), cex = 1.5)
     lines(map.times, roi.median.fitted.xTofts,
     col = "red", lwd = 5)
     lines(map.times, roi.median.fitted.Tofts,
     col = "blue", lwd = 2)
     text(x = 0.5 * max(map.times), 1 * max(cc.median,
     na.rm = TRUE), paste("R package version =",
     KAT.version), cex = 1.5)
     image(array(1:2, dim = c(1, 2)), col = palette(colorRampPalette(c("red",
     "blue"))(2)), zlim = c(0, 2), xaxt = "n",
     yaxt = "n", xlab = modeltype1, main = modeltype2,
     cex.lab = 1.5, cex.main = 1.5, cex.axis = 1.5)
     image(array(1:2, dim = c(1, 2)), col = palette(colorRampPalette(c("red",
     "blue"))(2)), zlim = c(0, 2), xaxt = "n",
     yaxt = "n", xlab = modeltype1, main = modeltype2,
     cex.lab = 1.5, cex.main = 1.5, cex.axis = 1.5,
     add = TRUE)
     image(map.AIC.compare, xlim = c(x_min/nx,
     x_max/nx), ylim = c(y_min/ny, y_max/ny),
     col = palette(colorRampPalette(c("black",
     "red", "blue"))(3)), main = "Summary of model discrimination analysis",
     cex.axis = 1.5, cex.lab = 1.5)
     image(map.AIC.compare, xlim = c(x_min/nx,
     x_max/nx), ylim = c(y_min/ny, y_max/ny),
     col = palette(colorRampPalette(c("black",
     "red", "blue"))(3)), main = "Summary of model discrimination analysis",
     cex.axis = 1.5, cex.lab = 1.5, add = TRUE)
     if (AIF.shift == "ARTERY") {
     plot(map.times, aif, ylim = c(0, max(aif,
     aif.shifted)), xlab = "min", ylab = "contrast agent",
     main = paste("Shifted Arterial Input Function (",
     format(param.est.whole.roi.xTofts$tlag *
     60, digits = 3), " sec)", sep = ""),
     type = "n", cex = 3, cex.axis = 1.5,
     cex.lab = 1.5)
     lines(map.times, aif, col = "red")
     lines(map.times, aif.shifted)
     legend(x = 0.55 * max(map.times), y = max(aif,
     aif.shifted), c("raw AIF", "shifted AIF"),
     c("red", "black"), cex = 2)
     }
     if (AIF.shift == "VEIN") {
     plot(map.times, aif, ylim = c(0, max(aif,
     aif.shifted)), xlab = "min", ylab = "contrast agent",
     main = paste("Shifted Venous Input Function (",
     format(param.est.whole.roi.xTofts$tlag *
     60, digits = 3), " sec)", sep = ""),
     type = "n", cex = 3, cex.axis = 1.5,
     cex.lab = 1.5)
     lines(map.times, aif, col = "blue")
     lines(map.times, aif.shifted)
     legend(x = 0.55 * max(map.times), y = max(aif,
     aif.shifted), c("raw VIF", "shifted VIF"),
     c("blue", "black"), cex = 2)
     }
     if (AIF.shift == "NONE") {
     plot(map.times, aif, ylim = c(0, max(aif)),
     xlab = "min", ylab = "contrast agent",
     main = "Vascular Input Function", type = "n",
     cex = 3, cex.axis = 1.5, cex.lab = 1.5)
     lines(map.times, aif)
     legend(x = 0.55 * max(map.times), y = max(aif),
     c("raw VIF"), c("black"), cex = 2)
     }
     dev.off()
     cat("image printed to pdf.", "\n")
     cat("---", "\n")
     }
     if (newplot == 1) {
     dev.new(width = 5.15, height = 4.4, xpos = 1500,
     ypos = 432)
     newplot <- 2
     }
     dev.set(7)
     plot(map.times, conc, xlab = "min", ylab = "contrast agent",
     ylim = c(-max(conc, na.rm = TRUE)/5, 1.4 *
     max(conc, na.rm = TRUE)), cex = 1.5, main = paste(paste("red=",
     modeltype1, ", blue=", modeltype2, " (",
     sep = ""), paste("x=", round(xx_old), ", y=",
     round(yy_old), ")", sep = ""), sep = ""))
     value_xTofts <- MAP[xx, yy]
     if (is.finite(value_xTofts) == FALSE)
     value_xTofts <- 0
     if (value_xTofts != 0) {
     if (est.per.voxel.tlag == TRUE)
     paramsxT <- c(map.KtransxT[xx, yy], map.kepxT[xx,
     yy], map.vbxT[xx, yy], map.tlagxT[xx,
     yy])
     if (est.per.voxel.tlag == FALSE)
     paramsxT <- c(map.KtransxT[xx, yy], map.kepxT[xx,
     yy], map.vbxT[xx, yy], param.est.whole.roi.xTofts$tlag)
     paramsT <- c(map.KtransT[xx, yy], map.kepT[xx,
     yy])
     roi.model <- roi.modelxT
     simulation <- roi.model(p = paramsxT, t = map.times,
     dt = diff(map.times), cp = aif)
     roi.model <- roi.modelT
     simulation_2 <- roi.model(p = paramsT, t = map.times,
     dt = diff(map.times), cp = aif)
     lines(map.times, simulation_2, col = "blue",
     lwd = 2)
     lines(map.times, simulation, col = "red",
     lwd = 2, lty = 2)
     roi.model <- roi.modelxT
     }
     text(max(map.times)/4.5, 1.4 * max(conc, na.rm = TRUE),
     "Fitted xTofts params")
     text(max(map.times)/4.5, 1.4 * max(conc, na.rm = TRUE) *
     0.93, paste(paste("Ktrans =", format(map.KtransxT[xx,
     yy], digits = 3)), "min^-1"))
     text(max(map.times)/4.5, 1.4 * max(conc, na.rm = TRUE) *
     0.86, paste("ve =", format(map.vexT[xx, yy],
     digits = 3)))
     text(max(map.times)/4.5, 1.4 * max(conc, na.rm = TRUE) *
     0.79, paste("vb =", format(map.vbxT[xx, yy],
     digits = 3)))
     if (est.per.voxel.tlag == TRUE)
     text(max(map.times)/4.5, 1.4 * max(conc,
     na.rm = TRUE) * 0.72, paste("tlag =", format(map.tlagxT[xx,
     yy], digits = 3)))
     if (newplot2 == 1) {
     dev.new(width = 5.15, height = 3, xpos = 1500,
     ypos = 864)
     newplot2 <- 2
     }
     else dev.set(8)
     if (value_xTofts != 0) {
     simulation_AIC <- simulation
     simulation_AIC_2 <- simulation_2
     if (length(simulation) == length(conc)) {
     plot(map.times, conc - simulation, xlab = "min",
     ylab = "predicated - measured", cex = 1.5,
     col = "red", main = paste("AIC(Tofts)=",
     map.AIC.T[xx, yy], " AIC(xTofts)=",
     map.AIC.xT[xx, yy], sep = ""))
     lines(locfit(conc - simulation ~ map.times,
     acri = "ici"), col = "red", lwd = 2)
     abline(h = 0, col = "black", lwd = 2)
     if (is.finite(simulation_2[1]) == TRUE) {
     points(map.times, conc - simulation_2,
     xlab = "min", cex = 1.5, col = "blue")
     lines(locfit(conc - simulation_2 ~ map.times,
     acri = "ici"), col = "blue", lwd = 2,
     lty = 2)
     }
     }
     }
     dev.set(4)
     }
     }
     if (file.original == TRUE) {
     ptm <- proc.time()[3]
     proc.time.total <- format((proc.time()[3] - ptm_total)/60,
     digits = 2)
     args <- list(as.character(file), as.character(results_file),
     as.character(method.optimization), show.rt.fits,
     as.character(param.for.avdt), range.map, cutoff.map,
     export.matlab, export.RData, verbose, show.errors,
     try.silent, fracGTzero, AIF.shift, slice, ID.visit,
     est.per.voxel.tlag)
     names(args) <- c("file", "resultsfile", "methodoptimization",
     "showrtfits", "paramforavdt", "rangemap", "cutoffmap",
     "exportmatlab", "exportRData", "verbose", "showerrors",
     "trysilent", "fracGTzero", "AIFshift", "slice",
     "IDvisit", "estpervoxeltlag")
     roiplotparams <- list(x_min, x_max, y_min, y_max,
     MAP_ul)
     names(roiplotparams) <- c("xmin", "xmax", "ymin",
     "ymax", "MAPul")
     dummy_data <- dcemri.data
     dcemri.data <- list(args, map_cc_slice, map_cc_roi,
     cc.median, map.times, aif, aif.shifted, mask.roi,
     map.KtransxT, map.KtransxT.cv, map.tlagxT,
     map.tlagxT.cv, map.kepxT, map.kepxT.cv, map.vbxT,
     map.vbxT.cv, map.vexT, map.OptimValuexT, map.fitfailuresxT,
     param.est.medianxT, roi.median.fitted.xTofts,
     param.est.whole.roi.xTofts, cv.whole.roi.xTofts,
     map.KtransT, map.KtransT.cv, map.kepT, map.kepT.cv,
     map.veT, map.OptimValueT, map.fitfailuresT,
     param.est.medianT, roi.median.fitted.Tofts,
     param.est.whole.roi.Tofts, proc.time.total,
     roiplotparams, KAT.version, map.AIC.xT, map.AIC.T,
     map.AIC.compare, nx, ny, nt, cc_fittedxT, cc_fittedT,
     p0.T, p0.xT, IRF.results, map.EF)
     names(dcemri.data) <- c("args", "cc", "ccroi",
     "ccmedian", "maptimes", "aif", "aifshifted",
     "maskroi", "mapKtransxT", "mapKtransxTcv",
     "maptlagxT", "maptlagxTcv", "mapkepxT", "mapkepxTcv",
     "mapvbxT", "mapvbxTcv", "mapvexT", "mapOptimValuexT",
     "mapfitfailuresxT", "paramestmedianxT", "roimedianfittedxTofts",
     "paramestwholeroixTofts", "cvwholeroixTofts",
     "mapKtransT", "mapKtransTcv", "mapkepT", "mapkepTcv",
     "mapveT", "mapOptimValueT", "mapfitfailuresT",
     "paramestmedianT", "roimedianfittedTofts",
     "paramestwholeroiTofts", "proctimetotal", "roiplotparams",
     "KATversion", "mapAICxT", "mapAICT", "mapAICcompare",
     "nx", "ny", "nt", "ccfittedxT", "ccfittedT",
     "p0T", "p0xT", "IRFresults", "mapEF")
     if (export.RData == TRUE) {
     cat("writing results to ", paste(results_file,
     ".RData", sep = ""), "...", sep = "", "\n")
     save(dcemri.data, file = paste(results_file,
     ".RData", sep = ""))
     }
     if (export.matlab == TRUE) {
     cat("writing results to ", paste(results_file,
     ".mat", sep = ""), "...", sep = "", "\n")
     writeMat(paste(results_file, ".mat", sep = ""),
     args = args, mapccslice = map_cc_slice, mapccroi = map_cc_roi,
     ccmedian = cc.median, maptimes = map.times,
     aif = aif, aifshifted = aif.shifted, maskroi = mask.roi,
     mapKtransxT = map.KtransxT, mapKtransxTcv = map.KtransxT.cv,
     maptlagxT = map.tlagxT, maptlagxTcv = map.tlagxT.cv,
     mapkepxT = map.kepxT, mapkepxTcv = map.kepxT.cv,
     mapvbxT = map.vbxT, mapvbxTcv = map.vbxT.cv,
     mapvexT = map.vexT, mapOptimValuexT = map.OptimValuexT,
     mapfitfailuresxT = map.fitfailuresxT, paramestmedianxT = param.est.medianxT,
     roimedianfittedxTofts = roi.median.fitted.xTofts,
     paramestwholeroixTofts = param.est.whole.roi.xTofts,
     cvwholeroixTofts = cv.whole.roi.xTofts, mapKtransT = map.KtransT,
     mapKtransTcv = map.KtransT.cv, mapkepT = map.kepT,
     mapkepTcv = map.kepT.cv, mapveT = map.veT,
     mapOptimValueT = map.OptimValueT, mapfitfailuresT = map.fitfailuresT,
     paramestmedianT = param.est.medianT, roimedianfittedTofts = roi.median.fitted.Tofts,
     paramestwholeroiTofts = param.est.whole.roi.Tofts,
     proctimetotal = proc.time.total, roiplotparams = roiplotparams,
     KATversion = KAT.version, mapAICxT = map.AIC.xT,
     mapAICT = map.AIC.T, mapAICcompare = map.AIC.compare,
     nx = nx, ny = ny, nt = nt, ccfittedxT = cc_fittedxT,
     ccfittedT = cc_fittedT, p0T = p0.T, p0xT = p0.xT,
     IRFresults = IRF.results, mapEF = map.EF)
     }
     dcemri.data <- dummy_data
     cat("..done in", format((proc.time()[3] - ptm)/60,
     digits = 2), "minutes.", "\n")
     if (export.matlab == FALSE)
     cat("Run KAT(filename.RData) to visualize results.",
     "\n")
     if (export.matlab == TRUE)
     cat("Run KAT(filename.RData) or KAT(filename.mat) to visualize results.",
     "\n")
     cat("--------", "\n")
     }
     }
     }
    }
    <bytecode: 0x564e10fbab70>
    <environment: namespace:KATforDCEMRI>
     --- function search by body ---
    Function KAT in namespace KATforDCEMRI has this body.
     ----------- END OF FAILURE REPORT --------------
    Error in if (class(cov) != "try-error") { : the condition has length > 1
    Calls: demo ... withVisible -> eval -> eval -> runme -> system.time -> KAT
    Timing stopped at: 1.425 0.122 2.282
    Execution halted
Flavor: r-devel-linux-x86_64-debian-gcc

Version: 1.0.1
Check: examples
Result: ERROR
    Running examples in ‘KATforDCEMRI-Ex.R’ failed
    The error most likely occurred in:
    
    > ### Name: KAT
    > ### Title: Kinetic Analysis Tool for DCE-MRI
    > ### Aliases: KAT
    > ### Keywords: kinetic DCEMRI
    >
    > ### ** Examples
    >
    > ## Create temporary directory for example code output files
    > temp_dir <- tempdir(check=FALSE)
    > ##
    > current_dir <- getwd()
    > setwd(temp_dir)
    > ##
    > ## Run example code
    > demo(KAT, ask=FALSE)
    
    
     demo(KAT)
     ---- ~~~
    
    > ## KATforDCEMRI: a Kinetic Analysis Tool for DCE-MRI
    > ## Copyright 2018 Genentech, Inc.
    > ##
    > ## For questions or comments, please contact
    > ## Gregory Z. Ferl, Ph.D.
    > ## Genentech, Inc.
    > ## Development Sciences
    > ## 1 DNA Way, Mail stop 463A
    > ## South San Francisco, CA, United States of America
    > ## E-mail: ferl.gregory@gene.com
    >
    > runme <- function(){
    + data(dcemri.data, package="KATforDCEMRI")
    +
    + ## dir.create("KATforDCEMRI_benchmark_test")
    + ## setwd("KATforDCEMRI_benchmark_test")
    +
    + attach(dcemri.data)
    +
    + ## SHRINK THE ROI MASK
    + maskROI[,,] <- 0
    + #maskROI[32:42,32:42,] <- 1
    + maskROI[34:36,34:36,] <- 1
    +
    + runtime1 <- system.time(KAT.checkData(file.name="KAT", vector.times=vectorTimes, map.CC=mapCC, mask.ROI=maskROI, vector.AIF=vectorAIF))
    + runtime2 <- system.time(KAT(file = "KAT.RData", results_file="KAT_benchmark_test-full", range.map=1.05, cutoff.map=0.95, AIF.shift="NONE", tlag.Tofts.on=FALSE, export.matlab=FALSE))
    +
    + ## runtime3 <- system.time(KAT.checkData(file.name="KATtrunc", vector.times=vectorTimes[1:44], map.CC=mapCC[,,,1:44], mask.ROI=maskROI, vector.AIF=vectorAIF[1:44]))
    + ## runtime4 <- system.time(KAT(file = "KATtrunc.RData", results_file="KAT_benchmark_test-truncated", range.map=1.05, cutoff.map=0.95))
    + detach(dcemri.data)
    +
    + ## runtime <- format(runtime1[3] + runtime2[3] + runtime3[3] + runtime4[3], digits=3)
    + runtime <- format(runtime1[3] + runtime2[3], digits=3)
    +
    + ## KAT.plot(F1="KAT_benchmark_test-full_slice1.RData", F2="KAT_benchmark_test-full_slice2.RData", F3="KAT_benchmark_test-truncated_slice1.RData", F4="KAT_benchmark_test-truncated_slice2.RData", export.matlab=FALSE)
    + KAT.plot(F1="KAT_benchmark_test-full_slice1.RData", F2="KAT_benchmark_test-full_slice2.RData", F3="KAT_benchmark_test-full_slice3.RData", F4="KAT_benchmark_test-full_slice4.RData", export.matlab=FALSE)
    +
    + load("KAT_benchmark_test-full_slice1.RData")
    + Ktrans_s1A <- as.numeric(format(dcemri.data$paramestmedianxT$Ktrans.median, digits=3))
    + kep_s1A <- as.numeric(format(dcemri.data$paramestmedianxT$kep.median, digits=3))
    + vb_s1A <- as.numeric(format(dcemri.data$paramestmedianxT$vb.median, digits=3))
    + cvKtrans_s1A <- as.numeric(format(100*sd(as.vector(dcemri.data$mapKtransxT), na.rm=TRUE)/Ktrans_s1A, digits=3))
    + cvkep_s1A <- as.numeric(format(100*sd(as.vector(dcemri.data$mapkepxT), na.rm=TRUE)/kep_s1A, digits=3))
    + cvvb_s1A <- as.numeric(format(100*sd(as.vector(dcemri.data$mapvbxT), na.rm=TRUE)/vb_s1A, digits=3))
    +
    + Ktrans_Ts1A <- as.numeric(format(dcemri.data$paramestmedianT$Ktrans.median, digits=3))
    + kep_Ts1A <- as.numeric(format(dcemri.data$paramestmedianT$kep.median, digits=3))
    + cvKtrans_Ts1A <- as.numeric(format(100*sd(as.vector(dcemri.data$mapKtransT), na.rm=TRUE)/Ktrans_Ts1A, digits=3))
    + cvkep_Ts1A <- as.numeric(format(100*sd(as.vector(dcemri.data$mapkepT), na.rm=TRUE)/kep_s1A, digits=3))
    +
    + load("KAT_benchmark_test-full_slice2.RData")
    + Ktrans_s2A <- as.numeric(format(dcemri.data$paramestmedianxT$Ktrans.median, digits=3))
    + kep_s2A <- as.numeric(format(dcemri.data$paramestmedianxT$kep.median, digits=3))
    + vb_s2A <- as.numeric(format(dcemri.data$paramestmedianxT$vb.median, digits=3))
    + cvKtrans_s2A <- as.numeric(format(100*sd(as.vector(dcemri.data$mapKtransxT), na.rm=TRUE)/Ktrans_s2A, digits=3))
    + cvkep_s2A <- as.numeric(format(100*sd(as.vector(dcemri.data$mapkepxT), na.rm=TRUE)/kep_s2A, digits=3))
    + cvvb_s2A <- as.numeric(format(100*sd(as.vector(dcemri.data$mapvbxT), na.rm=TRUE)/vb_s2A, digits=3))
    +
    + Ktrans_Ts2A <- as.numeric(format(dcemri.data$paramestmedianT$Ktrans.median, digits=3))
    + kep_Ts2A <- as.numeric(format(dcemri.data$paramestmedianT$kep.median, digits=3))
    + cvKtrans_Ts2A <- as.numeric(format(100*sd(as.vector(dcemri.data$mapKtransT), na.rm=TRUE)/Ktrans_Ts2A, digits=3))
    + cvkep_Ts2A <- as.numeric(format(100*sd(as.vector(dcemri.data$mapkepT), na.rm=TRUE)/kep_s2A, digits=3))
    +
    + ## load("KAT_benchmark_test-truncated_slice1.RData")
    + load("KAT_benchmark_test-full_slice3.RData")
    + Ktrans_s1B <- as.numeric(format(dcemri.data$paramestmedianxT$Ktrans.median, digits=3))
    + kep_s1B <- as.numeric(format(dcemri.data$paramestmedianxT$kep.median, digits=3))
    + vb_s1B <- as.numeric(format(dcemri.data$paramestmedianxT$vb.median, digits=3))
    + cvKtrans_s1B <- as.numeric(format(100*sd(as.vector(dcemri.data$mapKtransxT), na.rm=TRUE)/Ktrans_s1B, digits=3))
    + cvkep_s1B <- as.numeric(format(100*sd(as.vector(dcemri.data$mapkepxT), na.rm=TRUE)/kep_s1B, digits=3))
    + cvvb_s1B <- as.numeric(format(100*sd(as.vector(dcemri.data$mapvbxT), na.rm=TRUE)/vb_s1B, digits=3))
    +
    + Ktrans_Ts1B <- as.numeric(format(dcemri.data$paramestmedianT$Ktrans.median, digits=3))
    + kep_Ts1B <- as.numeric(format(dcemri.data$paramestmedianT$kep.median, digits=3))
    + cvKtrans_Ts1B <- as.numeric(format(100*sd(as.vector(dcemri.data$mapKtransT), na.rm=TRUE)/Ktrans_Ts1B, digits=3))
    + cvkep_Ts1B <- as.numeric(format(100*sd(as.vector(dcemri.data$mapkepT), na.rm=TRUE)/kep_s1B, digits=3))
    +
    + ## load("KAT_benchmark_test-truncated_slice2.RData")
    + load("KAT_benchmark_test-full_slice4.RData")
    + Ktrans_s2B <- as.numeric(format(dcemri.data$paramestmedianxT$Ktrans.median, digits=3))
    + kep_s2B <- as.numeric(format(dcemri.data$paramestmedianxT$kep.median, digits=3))
    + vb_s2B <- as.numeric(format(dcemri.data$paramestmedianxT$vb.median, digits=3))
    + cvKtrans_s2B <- as.numeric(format(100*sd(as.vector(dcemri.data$mapKtransxT), na.rm=TRUE)/Ktrans_s2B, digits=3))
    + cvkep_s2B <- as.numeric(format(100*sd(as.vector(dcemri.data$mapkepxT), na.rm=TRUE)/kep_s2B, digits=3))
    + cvvb_s2B <- as.numeric(format(100*sd(as.vector(dcemri.data$mapvbxT), na.rm=TRUE)/vb_s2B, digits=3))
    +
    + Ktrans_Ts2B <- as.numeric(format(dcemri.data$paramestmedianT$Ktrans.median, digits=3))
    + kep_Ts2B <- as.numeric(format(dcemri.data$paramestmedianT$kep.median, digits=3))
    + cvKtrans_Ts2B <- as.numeric(format(100*sd(as.vector(dcemri.data$mapKtransT), na.rm=TRUE)/Ktrans_Ts2B, digits=3))
    + cvkep_Ts2B <- as.numeric(format(100*sd(as.vector(dcemri.data$mapkepT), na.rm=TRUE)/kep_s2B, digits=3))
    +
    + pdf(file="KAT_demo-page1.pdf", height=11, width=8.5)
    +
    + plot(0:35, type = "n", xaxt="n", yaxt="n", bty="n", xlab = "", ylab = "")
    + text(5, 35, paste("KATforDCEMRI version ", dcemri.data$KATversion, " BENCHMARK TEST", sep=""), pos=4, font=2, col="red")
    + text(5, 33, paste("date:", date(), "\n"), pos=4)
    + text(5, 32, paste("Total Processing Time:", runtime, "seconds"), pos=4)
    + text(5, 29, paste("sysname/release:", Sys.info()[[1]], Sys.info()[[2]], "\n"), pos=4)
    + text(5, 28, paste("nodename:", Sys.info()[[4]], "\n"), pos=4)
    + text(5, 27, paste("user:", Sys.info()[[7]], "\n"), pos=4)
    + text(5, 25, "Estimated (inter-voxel %CV) versus True Parameter values (extended Tofts)", pos=4, font=2)
    + text(5, 23, "LOW NOISE DATA", pos=4)
    + text(5, 22, paste("Slice 1: Ktrans =", Ktrans_s1A, "1/min (", cvKtrans_s1A,"%) [true value = 0.22 1/min]", sep=""), pos=4)
    + text(5, 21, paste("Slice 1: kep =", kep_s1A, "1/min (", cvkep_s1A, "%) [true value = 1.1 1/min]", sep=""), pos=4)
    + text(5, 20, paste("Slice 1: vb =", vb_s1A, " (", cvvb_s1A, "%) [true value = 0]", sep=""), pos=4)
    + text(5, 18, paste("Slice 2: Ktrans =", Ktrans_s2A, "1/min (", cvKtrans_s2A,"%) [true value = 0.22 1/min]", sep=""), pos=4)
    + text(5, 17, paste("Slice 2: kep =", kep_s2A, "1/min (", cvkep_s2A, "%) [true value = 1.1 1/min]", sep=""), pos=4)
    + text(5, 16, paste("Slice 2: vb =", vb_s2A, " (", cvvb_s2A, "%) [true value = 0.05]", sep=""), pos=4)
    + text(5, 14, "HIGH NOISE DATA", pos=4)
    + text(5, 13, paste("Slice 3: Ktrans =", Ktrans_s1B, "1/min (", cvKtrans_s1B,"%) [true value = 0.22 1/min]", sep=""), pos=4)
    + text(5, 12, paste("Slice 3: kep =", kep_s1B, "1/min (", cvkep_s1B, "%) [true value = 1.1 1/min]", sep=""), pos=4)
    + text(5, 11, paste("Slice 3: vb =", vb_s1B, " (", cvvb_s1B, "%) [true value = 0]", sep=""), pos=4)
    + text(5, 9, paste("Slice 4: Ktrans =", Ktrans_s2B, "1/min (", cvKtrans_s2B,"%) [true value = 0.22 1/min]", sep=""), pos=4)
    + text(5, 8, paste("Slice 4: kep =", kep_s2B, "1/min (", cvkep_s2B, "%) [true value = 1.1 1/min]", sep=""), pos=4)
    + text(5, 7, paste("Slice 4: vb =", vb_s2B, " (", cvvb_s2B, "%) [true value = 0.05]", sep=""), pos=4)
    + dev.off()
    +
    +
    + pdf(file="KAT_demo-page2.pdf", height=11, width=8.5)
    +
    + plot(0:35, type = "n", xaxt="n", yaxt="n", bty="n", xlab = "", ylab = "")
    +
    + text(5, 25, "Estimated (inter-voxel %CV) versus True Parameter values (Tofts Model)", pos=4, font=2)
    + text(5, 23, "LOW NOISE DATA", pos=4)
    + text(5, 22, paste("Slice 1: Ktrans =", Ktrans_Ts1A, "1/min (", cvKtrans_Ts1A,"%) [true value = 0.22 1/min]", sep=""), pos=4)
    + text(5, 21, paste("Slice 1: kep =", kep_Ts1A, "1/min (", cvkep_Ts1A, "%) [true value = 1.1 1/min]", sep=""), pos=4)
    + text(5, 20, "Slice 1: vb = 0 (Fixed) [true value = 0]", pos=4)
    + text(5, 18, paste("Slice 2: Ktrans =", Ktrans_Ts2A, "1/min (", cvKtrans_Ts2A,"%) [true value = 0.22 1/min]", sep=""), pos=4)
    + text(5, 17, paste("Slice 2: kep =", kep_Ts2A, "1/min (", cvkep_Ts2A, "%) [true value = 1.1 1/min]", sep=""), pos=4)
    + text(5, 16, "Slice 2: vb = 0 (Fixed) [true value = 0.05]", pos=4)
    + text(5, 14, "HIGH NOISE DATA", pos=4)
    + text(5, 13, paste("Slice 3: Ktrans =", Ktrans_Ts1B, "1/min (", cvKtrans_Ts1B,"%) [true value = 0.22 1/min]", sep=""), pos=4)
    + text(5, 12, paste("Slice 3: kep =", kep_Ts1B, "1/min (", cvkep_Ts1B, "%) [true value = 1.1 1/min]", sep=""), pos=4)
    + text(5, 11, "Slice 3: vb = 0 (Fixed) [true value = 0]", pos=4)
    + text(5, 9, paste("Slice 4: Ktrans =", Ktrans_Ts2B, "1/min (", cvKtrans_Ts2B,"%) [true value = 0.22 1/min]", sep=""), pos=4)
    + text(5, 8, paste("Slice 4: kep =", kep_Ts2B, "1/min (", cvkep_Ts2B, "%) [true value = 1.1 1/min]", sep=""), pos=4)
    + text(5, 7, "Slice 4: vb = 0 (Fixed) [true value = 0.05]", pos=4)
    + dev.off()
    + }
    
    > runme()
    
    checking dimensions of vectors and arrays...
    
    length of vector.times is 89 with units of seconds
    length of vector.AIF is 89
    dimensions of map.CC array are 75 x 75 x 4 slices x 89 time points
    dimensions of mask.ROI array are 75 x 75 x 4 slices
    
    ...vector and array dimensions are okay.
    
    Saving data in a single R file...
    ...file saved as KAT.RData ...
    ...use the KAT() function to analyze data within this file.
    
    
    #########################################################################
    ##-------------- KINETIC ANALYSIS TOOL (KAT) FOR DCE-MRI --------------##
    #########################################################################
    ##---------------------- R package version 1.0 ----------------------##
    #########################################################################
    
    loading KAT.RData into R...
    ..done in 0.0015 minutes.
    --------
    ***** ROI DETECTED IN SLICE 1 *****
    --------
    extracting slice 1 for analysis...
    ..done in 0.0021 minutes.
    --------
    applying ROI mask to cc matrix...
    ..done in 0.00042 minutes.
    --------
    fitting xTofts and Tofts models to whole ROI data...
    ..done in 0.0039 minutes.
    --------
     ----------- FAILURE REPORT --------------
     --- failure: the condition has length > 1 ---
     --- srcref ---
    :
     --- package (from environment) ---
    KATforDCEMRI
     --- call from context ---
    KAT(file = "KAT.RData", results_file = "KAT_benchmark_test-full",
     range.map = 1.05, cutoff.map = 0.95, AIF.shift = "NONE",
     tlag.Tofts.on = FALSE, export.matlab = FALSE)
     --- call from argument ---
    if (class(cov) != "try-error") {
     sd <- try(sqrt(diag(cov)), silent = try.silent)
     if (class(sd) != "try-error") {
     cv <- sd/param_est * 100
     cv.whole.roi.xTofts <- as.numeric(format(cv, digits = 1))
     }
    }
     --- R stacktrace ---
    where 1 at /data/gannet/ripley/R/packages/tests-clang/KATforDCEMRI.Rcheck/KATforDCEMRI/demo/KAT.R#26: KAT(file = "KAT.RData", results_file = "KAT_benchmark_test-full",
     range.map = 1.05, cutoff.map = 0.95, AIF.shift = "NONE",
     tlag.Tofts.on = FALSE, export.matlab = FALSE)
    where 2 at /data/gannet/ripley/R/packages/tests-clang/KATforDCEMRI.Rcheck/KATforDCEMRI/demo/KAT.R#26: system.time(KAT(file = "KAT.RData", results_file = "KAT_benchmark_test-full",
     range.map = 1.05, cutoff.map = 0.95, AIF.shift = "NONE",
     tlag.Tofts.on = FALSE, export.matlab = FALSE))
    where 3 at /data/gannet/ripley/R/packages/tests-clang/KATforDCEMRI.Rcheck/KATforDCEMRI/demo/KAT.R#141: runme()
    where 4: eval(ei, envir)
    where 5: eval(ei, envir)
    where 6: withVisible(eval(ei, envir))
    where 7: source(available, echo = echo, max.deparse.length = Inf, keep.source = TRUE,
     encoding = encoding)
    where 8: demo(KAT, ask = FALSE)
    
     --- value of length: 2 type: logical ---
    [1] TRUE TRUE
     --- function from context ---
    function (file = "concatenate.KAT.with.KAT.checkData.RData",
     results_file = "my_results", method.optimization = "L-BFGS-B",
     show.rt.fits = FALSE, param.for.avdt = "Ktrans", range.map = 1.5,
     cutoff.map = 0.85, export.matlab = TRUE, export.RData = TRUE,
     verbose = FALSE, show.errors = FALSE, try.silent = TRUE,
     fracGTzero = 0.75, AIF.shift = "", Force.AIF.peak = FALSE,
     tlag.Tofts.on = FALSE, est.per.voxel.tlag = FALSE, ...)
    {
     lo <- 0
     options(show.error.messsages = show.errors)
     ftype <- strsplit(file, split = "a")[[1]]
     ftype <- ftype[length(ftype) - 1]
     ftype <- strsplit(ftype, split = "")[[1]]
     ftype <- ftype[length(ftype)]
     if (ftype == "m")
     file.format <- "matlab"
     if (ftype == "D")
     file.format <- "RData"
     file_short <- file
     KAT.version <- "1.0"
     ptm_total <- proc.time()[3]
     modeltype1 <- "xTofts"
     modeltype2 <- "Tofts"
     funcz <- function(x) {
     as.numeric(as.character(x))
     }
     aif.shift.func <- function(t, cp, time_shift) {
     x <- cbind(t, cp)
     tmax <- subset(x[, 1], x[, 2] == max(x[, 2]))
     if (AIF.shift == "ARTERY")
     tmax <- tmax + time_shift
     if (AIF.shift == "VEIN")
     tmax <- tmax - time_shift
     cpFUNC <- approxfun(t, cp, rule = 2)
     if (AIF.shift == "ARTERY")
     tshift <- t - time_shift
     if (AIF.shift == "VEIN")
     tshift <- t + time_shift
     cp.shift <- cpFUNC(tshift)
     if (Force.AIF.peak == TRUE)
     cp.shift[cp.shift == max(cp.shift)] <- max(cp)
     return(cp.shift)
     }
     roi.modelT <- function(p, t, dt, cp, zing = 0) {
     if (zing == 0) {
     Ktrans <- p[1]
     kep <- p[2]
     if (tlag.Tofts.on == TRUE) {
     if (fix.tlag != TRUE & AIF.shift != "NONE")
     time_shift <- p[3]
     if (fix.tlag == TRUE & AIF.shift != "NONE")
     time_shift <- param.est.whole.roi.Tofts$tlag
     }
     if (tlag.Tofts.on == FALSE) {
     time_shift <- param.est.whole.roi.xTofts$tlag
     }
     if (AIF.shift == "ARTERY" | AIF.shift == "VEIN")
     cp <- aif.shift.func(t, cp, time_shift)
     f <- Ktrans * exp(kep * t) * cp
     int <- c(0, cumsum(dt * (tail(f, -1) + head(f, -1)))/2)
     ct <- exp(-kep * t) * int
     return(ct)
     }
     if (zing == 1)
     return("modelT")
     }
     roi.modelxT <- function(p, t, dt, cp, zing = 0) {
     if (zing == 0) {
     Ktrans <- p[1]
     kep <- p[2]
     vb <- p[3]
     if (fix.tlag != TRUE & AIF.shift != "NONE")
     time_shift <- p[4]
     if (est.per.voxel.tlag == FALSE) {
     if (fix.tlag == TRUE & AIF.shift != "NONE")
     time_shift <- param.est.whole.roi.xTofts$tlag
     }
     if (est.per.voxel.tlag == TRUE) {
     if (fix.tlag == TRUE & AIF.shift != "NONE")
     time_shift <- p[4]
     }
     if (AIF.shift == "ARTERY" | AIF.shift == "VEIN")
     cp <- aif.shift.func(t, cp, time_shift)
     f <- Ktrans * exp(kep * t) * cp
     int <- c(0, cumsum(dt * (tail(f, -1) + head(f, -1)))/2)
     ct <- exp(-kep * t) * int
     ct <- ct + vb * cp
     return(ct)
     }
     if (zing == 1)
     return("modelxT")
     }
     calch <- function(u, y, TIME_trunc) {
     locfit_y <- preplot(y, newdata = 0:max(TIME_trunc))
     y_smooth <- locfit_y$fit
     u <- u[match(u[u == max(u)], u):length(u)]
     y_smooth <- y_smooth[match(u[u == max(u)], u):length(u)]
     n <- length(u)
     A <- matrix(0, nrow = n, ncol = n)
     ind <- row(A) - col(A)
     ind[ind < 0] <- (-1)
     ind <- ind + 2
     A <- matrix(c(0, u)[ind], nrow = n, ncol = n)
     h <- solve(A, y_smooth)
     h.time.vector <- (1:length(h)) - 1
     out <- list(h.time.vector, h)
     names(out) <- c("t", "IRF")
     return(out)
     }
     calchFUNC <- function(vector.times, AIF, map_cc_slice, correct.vp = TRUE,
     alpha.AIF = c(0.1, 0.5), vp.nom = 0.1, kep.nom = 0.5) {
     AUMC <- function(AUMC.median, h.median, irf_time_vec,
     r) {
     AUMC.median <- AUMC.median + 0.5 * (h.median[r] *
     irf_time_vec[r] + h.median[r + 1] * irf_time_vec[r +
     1])
     }
     artery_data <- data.frame(vector.times * 60, AIF)
     names(artery_data) <- c("TIME", "ARTERY")
     data_artery_peak <- subset(artery_data, artery_data$ARTERY ==
     max(artery_data$ARTERY))
     data_remove_artery_prepeak <- subset(artery_data, artery_data$TIME >=
     data_artery_peak$TIME)
     frames_to_peak <- length(artery_data[, 1]) - length(data_remove_artery_prepeak[,
     1]) + 1
     TIME <- data_remove_artery_prepeak$TIME
     ARTERY <- data_remove_artery_prepeak$ARTERY
     TIME_trunc <- TIME[seq(1, length(TIME) - 1, by = 1)]
     TIME_trunc <- TIME_trunc - TIME_trunc[1]
     ARTERY_trunc <- ARTERY[seq(1, length(ARTERY) - 1, by = 1)]
     ARTERY_smooth <- locfit.robust(ARTERY_trunc ~ TIME_trunc,
     acri = "cp", alpha = alpha.AIF)
     AIF_smooth <- ARTERY_smooth
     locfit_u <- preplot(AIF_smooth, newdata = 0:max(TIME_trunc))
     u_smooth <- locfit_u$fit
     Tmax <- max(TIME_trunc)
     TUMOR.median <- map_cc_slice
     TUMOR.median <- TUMOR.median[seq(frames_to_peak, length(TUMOR.median),
     by = 1)]
     if (vp.nom > 0)
     TUMOR.median_corr <- TUMOR.median - vp.nom * ARTERY
     if (vp.nom <= 0)
     TUMOR.median_corr <- TUMOR.median
     TUMOR.median_corr_shifted <- TUMOR.median_corr[seq(2,
     length(TUMOR.median_corr), by = 1)]
     TUMOR.median_smooth <- locfit.robust(TUMOR.median_corr_shifted ~
     TIME_trunc, acri = "cp")
     calch.out <- calch(u_smooth, TUMOR.median_smooth, TIME_trunc)
     h.median <- calch.out$IRF
     irf_time_vec <- calch.out$t
     n <- length(h.median)
     AUC.median <- 0
     AUMC.median <- 0
     for (r in 1:(n - 1)) {
     h_sum <- h.median[r] + h.median[r + 1]
     t_sum <- irf_time_vec[r] + irf_time_vec[r + 1]
     AUC.median <- AUC.median + 0.5 * h_sum
     AUMC.median <- AUMC(AUMC.median, h.median, irf_time_vec,
     r)
     }
     AUCMRT.median <- AUC.median/(AUMC.median/AUC.median) *
     60
     if (kep.nom > 0) {
     t_scan <- max(TIME_trunc)/60
     ve_trunc_error <- 1 - exp(-kep.nom * t_scan)
     Ktrans_trunc_error <- (1 - exp(-kep.nom * t_scan))^2/(1 -
     (1 + kep.nom * t_scan) * exp(-kep.nom * t_scan))
     AUC.median <- AUC.median/ve_trunc_error
     AUCMRT.median <- AUCMRT.median/Ktrans_trunc_error
     }
     irf_time_vec <- irf_time_vec/60
     out <- list(AUC.median, AUCMRT.median, h.median, irf_time_vec)
     names(out) <- c("AUCh", "AUChMRTh", "IRF", "t")
     return(out)
     }
     Obj_roi <- function(p, model, t, dt, cp, roi) {
     sum((model(p, t, dt, cp) - roi)^2)
     }
     map_cc_slice <- NULL
     map_cc_roi <- NULL
     aif <- NULL
     aif.shifted <- NULL
     map.times <- NULL
     map.KtransxT <- NULL
     map.kepxT <- NULL
     map.vexT <- NULL
     map.vbxT <- NULL
     map.tlagxT <- NULL
     map.fitfailuresxT <- NULL
     map.KtransT <- NULL
     map.kepT <- NULL
     map.veT <- NULL
     map.fitfailuresT <- NULL
     mask.roi <- NULL
     nx <- NULL
     ny <- NULL
     nt <- NULL
     map.KtransxT <- NULL
     map.kepxT <- NULL
     map.vexT <- NULL
     map.vbxT <- NULL
     map.AIC.compare <- NULL
     map.AIC.T <- NULL
     map.EF <- NULL
     map.AIC.xT <- NULL
     roi.median.fitted.Tofts <- NULL
     param.est.whole.roi.Tofts <- NULL
     roi.median.fitted.xTofts <- NULL
     param.est.whole.roi.xTofts <- NULL
     cv.whole.roi.xTofts <- NULL
     cat("\n")
     cat("#########################################################################",
     "\n")
     cat("##-------------- KINETIC ANALYSIS TOOL (KAT) FOR DCE-MRI --------------##",
     "\n")
     cat("#########################################################################",
     "\n")
     cat("##---------------------- R package version", KAT.version,
     "----------------------##", "\n")
     cat("#########################################################################",
     "\n")
     cat("\n")
     filea <- strsplit(file, split = "/")[[1]]
     fileb <- filea[length(filea)]
     if (file != "concatenate.KAT.with.KAT.checkData.RData")
     cat("loading", fileb, "into R...", "\n")
     ptm <- proc.time()[3]
     if (file.format == "matlab") {
     mat_data <- readMat(file)
     if (length(names(mat_data)) < 10)
     file.original <- TRUE
     if (length(names(mat_data)) >= 10)
     file.original <- FALSE
     }
     if (file.format == "RData") {
     if (file != "concatenate.KAT.with.KAT.checkData.RData")
     load(file)
     if (length(names(dcemri.data)) < 10) {
     file.original <- TRUE
     ROIcounter <- apply(dcemri.data$maskROI, 3, max)
     results_file_temp <- results_file
     }
     if (length(names(dcemri.data)) >= 10) {
     file.original <- FALSE
     ROIcounter <- 1
     }
     }
     if (file != "concatenate.KAT.with.KAT.checkData.RData") {
     cat("..done in", format((proc.time()[3] - ptm)/60, digits = 2),
     "minutes.", "\n")
     cat("--------", "\n")
     }
     for (slicenumber in 1:(length(ROIcounter))) {
     if (ROIcounter[slicenumber] == 1) {
     if (file.original == TRUE) {
     slice <- slicenumber
     cat("***** ROI DETECTED IN SLICE", slice, " *****",
     "\n")
     cat("--------", "\n")
     }
     if (file.original == TRUE) {
     if (AIF.shift != "VEIN" & AIF.shift != "ARTERY" &
     AIF.shift != "NONE")
     stop("You must specify the argument AIF.shift argument as VEIN, ARTERY or NONE, indicating that the AIF you are using is based on data from a vein or artery or NONE if tlag should be set to 0. This will ensure that the time lag parameter in the Tofts and xTofts models has the appropriate inital value and is bounded correctly; either -Inf to 0 (for VEIN) or 0 to Inf (for ARTERY).")
     filenameTag <- paste("_slice", slice, sep = "")
     results_file <- paste(results_file_temp, filenameTag,
     sep = "")
     roi.model <- roi.modelxT
     if (slice == "" || slice < 0)
     stop("The slice argument has not been properly specified; slice=``slice number''")
     cat("extracting slice", slice, "for analysis...",
     "\n")
     ptm <- proc.time()[3]
     if (file.format == "matlab") {
     map.times <- as.vector(mat_data$map[[4]]/60)
     map_cc <- mat_data$map[[3]]
     map_cc_slice <- map_cc[, , slice, ]
     if (length(map.times) != length(map_cc_slice[1,
     1, ]))
     stop("The length of the time vector does not match the length of the contrast agent concentration vector")
     }
     if (file.format == "RData") {
     map.times <- as.vector(dcemri.data$vectorTimes/60)
     map_cc <- dcemri.data$mapCC
     map_cc_slice <- map_cc[, , slice, ]
     if (length(map.times) != length(map_cc_slice[1,
     1, ]))
     stop("The length of the time vector does not match the length of the contrast agent concentration vector")
     }
     nt <- length(map_cc_slice[1, 1, ])
     ny <- length(map_cc_slice[, 1, 1])
     nx <- length(map_cc_slice[1, , 1])
     ccTEMP <- rep(0, prod(dim(map_cc_slice)))
     dim(ccTEMP) <- dim(map_cc_slice)[c(2, 1, 3)]
     for (i in 1:nt) ccTEMP[, , i] <- rot90(map_cc_slice[,
     , i], 3)
     map_cc_slice <- ccTEMP
     if (file.format == "matlab")
     mask.roi <- mat_data$mask[[1]][, , slice]
     if (file.format == "RData")
     mask.roi <- dcemri.data$mask[, , slice]
     mask.roi <- rot90(mask.roi, 3)
     if (max(mask.roi) != 1)
     stop("Your ROI mask is either composed entirely of zeroes or contains nonnumeric elements; voxels within the ROI should have a value of ``1'' and all other voxels should have a value of ``0''.")
     if (file.format == "matlab")
     aif <- as.vector(mat_data$aif)
     if (file.format == "RData")
     aif <- as.vector(dcemri.data$vectorAIF)
     if (file.format == "matlab")
     rm(mat_data)
     cat("..done in", format((proc.time()[3] - ptm)/60,
     digits = 2), "minutes.", "\n")
     cat("--------", "\n")
     cat("applying ROI mask to cc matrix...", "\n")
     ptm <- proc.time()[3]
     map_cc_roi <- map_cc_slice
     mask.roi.temp <- mask.roi
     mask.roi.temp[mask.roi.temp == 0] <- NA
     for (z in 1:nt) map_cc_roi[, , z] <- map_cc_roi[,
     , z] * mask.roi.temp
     cat("..done in", format((proc.time()[3] - ptm)/60,
     digits = 2), "minutes.", "\n")
     cat("--------", "\n")
     ptm <- proc.time()[3]
     cc.median <- seq(1:nt)
     for (t in 1:nt) cc.median[t] <- median(map_cc_roi[,
     , t], na.rm = TRUE)
     cat("fitting", modeltype1, "and", modeltype2,
     "models to whole ROI data...", "\n")
     IRF.out <- calchFUNC(map.times, aif, cc.median)
     AUCcorrnom <- IRF.out$AUCh
     AUCMRTcorrnom <- IRF.out$AUChMRTh
     AUCMRTcorrnom.divby.AUCcorrnom <- IRF.out$AUChMRTh/IRF.out$AUCh
     if (tlag.Tofts.on == TRUE & AIF.shift == "VEIN") {
     p0.T.median <- c(IRF.out$AUChMRTh, IRF.out$AUChMRTh/IRF.out$AUCh,
     0.05)
     lower.wholeT <- c(lo, lo, 0)
     upper.wholeT <- c(Inf, Inf, Inf)
     }
     if (tlag.Tofts.on == TRUE & AIF.shift == "ARTERY") {
     p0.T.median <- c(IRF.out$AUChMRTh, IRF.out$AUChMRTh/IRF.out$AUCh,
     0.05)
     lower.wholeT <- c(lo, lo, 0)
     upper.wholeT <- c(Inf, Inf, Inf)
     }
     if (tlag.Tofts.on == FALSE) {
     p0.T.median <- c(IRF.out$AUChMRTh, IRF.out$AUChMRTh/IRF.out$AUCh)
     lower.wholeT <- c(lo, lo)
     upper.wholeT <- c(Inf, Inf)
     }
     if (AIF.shift == "VEIN") {
     p0.xT.median <- c(IRF.out$AUChMRTh, IRF.out$AUChMRTh/IRF.out$AUCh,
     0.05, 0.05)
     lower.wholexT <- c(lo, lo, lo, 0)
     upper.wholexT <- c(Inf, Inf, Inf, Inf)
     }
     if (AIF.shift == "ARTERY") {
     p0.xT.median <- c(IRF.out$AUChMRTh, IRF.out$AUChMRTh/IRF.out$AUCh,
     0.05, 0.05)
     lower.wholexT <- c(lo, lo, lo, 0)
     upper.wholexT <- c(Inf, Inf, Inf, Inf)
     }
     if (AIF.shift == "NONE") {
     p0.xT.median <- c(IRF.out$AUChMRTh, IRF.out$AUChMRTh/IRF.out$AUCh,
     0.05)
     lower.wholexT <- c(lo, lo, lo)
     upper.wholexT <- c(Inf, Inf, Inf)
     }
     SAAMII <- FALSE
     if (SAAMII == TRUE) {
     hw.x <- cbind(format(map.times, digits = 3),
     format(aif, digits = 3), format(cc.median,
     digits = 3))
     write.table("DATA", file = paste("slice", slice,
     "_forSAAMII", sep = ""), quote = FALSE, row.names = FALSE,
     col.names = FALSE)
     write.table("(SD 1)", file = paste("slice",
     slice, "_forSAAMII", sep = ""), quote = FALSE,
     row.names = FALSE, append = TRUE, col.names = FALSE)
     write.table(hw.x, file = paste("slice", slice,
     "_forSAAMII", sep = ""), quote = FALSE, row.names = FALSE,
     append = TRUE, col.names = c("t", paste("AIF_s",
     slice, sep = ""), paste("CC_s", slice,
     sep = "")))
     write.table("END", file = paste("slice", slice,
     "_forSAAMII", sep = ""), quote = FALSE, row.names = FALSE,
     append = TRUE, col.names = FALSE)
     }
     roi <- cc.median
     t <- map.times
     fix.tlag <- FALSE
     roi.model <- roi.modelxT
     if (method.optimization == "L-BFGS-B")
     fit.roi.median.xTofts <- try(optim(p0.xT.median,
     Obj_roi, model = roi.model, t = map.times,
     dt = diff(map.times), cp = aif, roi = roi,
     method = "L-BFGS-B", lower = lower.wholexT,
     upper = upper.wholexT, hessian = TRUE), silent = try.silent)
     if (method.optimization != "L-BFGS-B")
     fit.roi.median.xTofts <- try(optim(p0.xT.median,
     Obj_roi, model = roi.model, t = map.times,
     dt = diff(map.times), cp = aif, roi = roi,
     method = method.optimization, hessian = TRUE),
     silent = try.silent)
     if (class(fit.roi.median.xTofts) != "try-error") {
     if (fit.roi.median.xTofts$convergence == 0) {
     roi.median.fitted.xTofts <- roi.model(p = fit.roi.median.xTofts$par,
     t = map.times, dt = diff(map.times), cp = aif)
     params.xTofts <- fit.roi.median.xTofts$par
     param.est.whole.roi.xTofts <- list(fit.roi.median.xTofts$par[1],
     fit.roi.median.xTofts$par[2], fit.roi.median.xTofts$par[3],
     fit.roi.median.xTofts$par[4])
     names(param.est.whole.roi.xTofts) <- c("Ktrans",
     "kep", "vb", "tlag")
     }
     }
     if (class(fit.roi.median.xTofts) == "try-error") {
     cat("A problem occured when fitting the extended Tofts model to median intensity/concentration data across the ROI. Optimization algorithm did not converge. This may occur when a large fraction of voxels within the ROI are non-enhancing, as these are not excluded from analysis of the median intensity/concentration profile",
     "\n")
     }
     if (class(fit.roi.median.xTofts) != "try-error") {
     if (fit.roi.median.xTofts$convergence != 0) {
     cat("A problem occured when fitting the Tofts model to median intensity/concentration data across the ROI (convergence code not equal to zero). Optimization algorithm did not converge. This may occur when a large fraction of voxels within the ROI are non-enhancing, as these are not excluded from analysis of the median intensity/concentration profile",
     "\n")
     }
     }
     roi.model <- roi.modelT
     if (method.optimization == "L-BFGS-B")
     fit.roi.median.Tofts <- try(optim(p0.T.median,
     Obj_roi, model = roi.model, t = map.times,
     dt = diff(map.times), cp = aif, roi = roi,
     method = "L-BFGS-B", lower = lower.wholeT,
     upper = upper.wholeT, hessian = TRUE), silent = try.silent)
     if (method.optimization != "L-BFGS-B")
     fit.roi.median.Tofts <- try(optim(p0.T.median,
     Obj_roi, model = roi.model, t = map.times,
     dt = diff(map.times), cp = aif, roi = roi,
     method = method.optimization, hessian = TRUE),
     silent = try.silent)
     if (class(fit.roi.median.Tofts) != "try-error") {
     if (fit.roi.median.Tofts$convergence == 0) {
     roi.median.fitted.Tofts <- roi.model(p = fit.roi.median.Tofts$par,
     t = map.times, dt = diff(map.times), cp = aif)
     if (tlag.Tofts.on == FALSE) {
     param.est.whole.roi.Tofts <- list(fit.roi.median.Tofts$par[1],
     fit.roi.median.Tofts$par[2])
     names(param.est.whole.roi.Tofts) <- c("Ktrans",
     "kep")
     }
     if (tlag.Tofts.on == TRUE) {
     param.est.whole.roi.Tofts <- list(fit.roi.median.Tofts$par[1],
     fit.roi.median.Tofts$par[2], fit.roi.median.Tofts$par[3])
     names(param.est.whole.roi.Tofts) <- c("Ktrans",
     "kep", "tlag")
     }
     }
     }
     if (class(fit.roi.median.Tofts) == "try-error") {
     cat("A problem occured when fitting the Tofts model to median intensity/concentration data across the ROI (a try-error occured). Optimization algorithm did not converge. This may occur when a large fraction of voxels within the ROI are non-enhancing, as these are not excluded from analysis of the median intensity/concentration profile",
     "\n")
     }
     if (class(fit.roi.median.Tofts) != "try-error") {
     if (fit.roi.median.Tofts$convergence != 0) {
     cat("A problem occured when fitting the Tofts model to median intensity/concentration data across the ROI (convergence code not equal to zero). Optimization algorithm did not converge. This may occur when a large fraction of voxels within the ROI are non-enhancing, as these are not excluded from analysis of the median intensity/concentration profile",
     "\n")
     }
     }
     roi.model <- roi.modelxT
     IRF.out <- calchFUNC(vector.times = map.times,
     AIF = aif, map_cc_slice = cc.median, vp.nom = param.est.whole.roi.xTofts$vb,
     kep.nom = param.est.whole.roi.xTofts$kep)
     p0.T <- c(IRF.out$AUChMRTh, IRF.out$AUChMRTh/IRF.out$AUCh)
     names(p0.T) <- c("Ktrans", "kep")
     if (est.per.voxel.tlag == FALSE) {
     p0.xT <- c(IRF.out$AUChMRTh, IRF.out$AUChMRTh/IRF.out$AUCh,
     param.est.whole.roi.xTofts$vb)
     names(p0.xT) <- c("Ktrans", "kep", "vb")
     }
     if (est.per.voxel.tlag == TRUE) {
     p0.xT <- c(IRF.out$AUChMRTh, IRF.out$AUChMRTh/IRF.out$AUCh,
     param.est.whole.roi.xTofts$vb, 0.05)
     names(p0.xT) <- c("Ktrans", "kep", "vb", "tlag")
     }
     AUCcorr <- IRF.out$AUCh
     AUCMRTcorr <- IRF.out$AUChMRTh
     AUCMRTcorr.divby.AUCcorr <- IRF.out$AUChMRTh/IRF.out$AUCh
     IRF.results <- c(AUCcorrnom, AUCMRTcorrnom, AUCMRTcorrnom.divby.AUCcorrnom,
     AUCcorr, AUCMRTcorr, AUCMRTcorr.divby.AUCcorr)
     names(IRF.results) <- c("AUCcorrnom(ve)", "AUCMRTcorrnom(Ktrans)",
     "AUCMRTcorrnom.divby.AUCcorrnom(kep)", "AUCcorr(ve)",
     "AUCMRTcorr(Ktrans)", "AUCMRTcorr.divby.AUCcorr(kep)")
     cat("..done in", format((proc.time()[3] - ptm)/60,
     digits = 2), "minutes.", "\n")
     cat("--------", "\n")
     roi.model <- roi.modelxT
     if (class(fit.roi.median.xTofts) != "try-error") {
     if (fit.roi.median.xTofts$convergence == 0) {
     RSS <- sum((roi.model(p = fit.roi.median.xTofts$par,
     t = map.times, dt = diff(map.times), cp = aif) -
     roi)^2)
     param_est <- fit.roi.median.xTofts$par
     nD <- nt
     nP <- length(param_est)
     hess <- fit.roi.median.xTofts$hessian
     cov <- try((nP * RSS/(nD - nP)) * solve(hess),
     silent = try.silent)
     cv <- param_est
     cv[] <- NA
     if (class(cov) != "try-error") {
     sd <- try(sqrt(diag(cov)), silent = try.silent)
     if (class(sd) != "try-error") {
     cv <- sd/param_est * 100
     cv.whole.roi.xTofts <- as.numeric(format(cv,
     digits = 1))
     }
     }
     param.roi <- as.numeric(format(param_est,
     digits = 3))
     }
     }
     if (AIF.shift == "ARTERY" | AIF.shift == "VEIN")
     aif.shifted <- aif.shift.func(map.times, aif,
     param.est.whole.roi.xTofts$tlag)
     if (show.rt.fits == TRUE) {
     if (AIF.shift == "ARTERY") {
     dev.new(width = 6, height = 6, xpos = 1500,
     ypos = 0)
     plot(map.times, aif, ylim = c(0, max(aif,
     aif.shifted)), xlab = "min", ylab = "contrast agent",
     main = paste("Shifted Arterial Input Function (",
     format(param.est.whole.roi.xTofts$tlag *
     60, digits = 3), " sec)", sep = ""),
     type = "n")
     lines(map.times, aif, col = "red")
     lines(map.times, aif.shifted)
     legend(x = 0.55 * max(map.times), y = max(aif,
     aif.shifted), c("raw AIF", "shifted AIF"),
     c("red", "black"))
     }
     if (AIF.shift == "VEIN") {
     dev.new(width = 6, height = 6, xpos = 1500,
     ypos = 0)
     plot(map.times, aif, ylim = c(0, max(aif,
     aif.shifted)), xlab = "min", ylab = "contrast agent",
     main = paste("Shifted Venous Input Function (",
     format(param.est.whole.roi.xTofts$tlag *
     60, digits = 3), " sec)", sep = ""),
     type = "n")
     lines(map.times, aif, col = "blue")
     lines(map.times, aif.shifted)
     legend(x = 0.55 * max(map.times), y = max(aif,
     aif.shifted), c("raw VIF", "shifted VIF"),
     c("blue", "black"))
     }
     if (AIF.shift == "NONE") {
     dev.new(width = 6, height = 6, xpos = 1500,
     ypos = 0)
     plot(map.times, aif, ylim = c(0, max(aif)),
     xlab = "min", ylab = "contrast agent",
     main = "Vascular Input Function", type = "n")
     lines(map.times, aif)
     legend(x = 0.55 * max(map.times), y = max(aif),
     c("raw VIF"), c("black"))
     }
     }
     if (show.rt.fits == TRUE) {
     dev.new(width = 6, height = 6, xpos = 1500,
     ypos = 575)
     plot(map.times, roi, xlab = "min", ylab = "contrast agent",
     main = "median contrast agent conc; Tofts=blue, xTofts=red",
     cex = 2)
     text(x = 0.65 * max(map.times), y = 0.4 * max(roi,
     na.rm = TRUE), labels = "EXTENDED TOFTS PARAMS",
     cex = 1)
     if (class(fit.roi.median.xTofts) != "try-error") {
     if (fit.roi.median.xTofts$convergence ==
     0) {
     text(x = 0.65 * max(map.times), y = 0.35 *
     max(roi, na.rm = TRUE), labels = paste(expression(K^trans),
     " = ", format(param.est.whole.roi.xTofts$Ktrans,
     digits = 3), " 1/min (", cv.whole.roi.xTofts[1],
     "%)", sep = ""), cex = 1)
     text(x = 0.65 * max(map.times), y = 0.3 *
     max(roi, na.rm = TRUE), labels = paste(expression(k_ep),
     " = ", format(param.est.whole.roi.xTofts$kep,
     digits = 3), " 1/min (", cv.whole.roi.xTofts[2],
     "%)", sep = ""), cex = 1)
     text(x = 0.65 * max(map.times), y = 0.25 *
     max(roi, na.rm = TRUE), labels = paste(expression(v_b),
     " = ", format(param.est.whole.roi.xTofts$vb,
     digits = 3), " dimensionless (", cv.whole.roi.xTofts[3],
     "%)", sep = ""), cex = 1)
     if (AIF.shift != "NONE")
     text(x = 0.65 * max(map.times), y = 0.2 *
     max(roi, na.rm = TRUE), labels = paste(expression(t_lag),
     " = ", format(60 * param.est.whole.roi.xTofts$tlag,
     digits = 3), " sec (", cv.whole.roi.xTofts[4],
     "%)", sep = ""), cex = 1)
     lines(map.times, roi.median.fitted.xTofts,
     col = "red", lwd = 5)
     }
     }
     if (class(fit.roi.median.Tofts) != "try-error") {
     if (fit.roi.median.Tofts$convergence == 0) {
     lines(map.times, roi.median.fitted.Tofts,
     col = "blue", lwd = 2)
     }
     }
     }
     }
     if (file.original == FALSE) {
     fix.tlag <- TRUE
     if (file.format == "matlab") {
     mat_data <- readMat(file)
     args <- mat_data$args[, , 1]
     results_file <- args$resultsfile
     ID.visit <- args$IDvisit
     slice <- args$slice
     method.optimization <- args$methodoptimization
     show.rt.fits <- args$showrtfits
     param.for.avdt <- param.for.avdt
     range.map <- args$rangemap
     cutoff.map <- args$cutoffmap
     lo <- args$lo
     est.per.voxel.tlag <- args$estpervoxeltlag
     p0.xT <- mat_data$p0xT
     p0.T <- mat_data$p0T
     AIF.shift <- args$AIFshift
     nx <- mat_data$nx
     ny <- mat_data$ny
     nt <- mat_data$nt
     map_cc_slice <- mat_data$cc
     map_cc_roi <- mat_data$ccroi
     map.times <- mat_data$maptimes
     aif <- mat_data$aif
     aif.shifted <- mat_data$aifshifted
     map.KtransxT <- mat_data$mapKtransxT
     map.tlagxT <- mat_data$maptlagxT
     map.kepxT <- mat_data$mapkepxT
     map.vexT <- mat_data$mapvexT
     map.vbxT <- mat_data$mapvbxT
     map.KtransT.cv <- mat_data$mapKtransTcv
     map.kepT.cv <- mat_data$mapkepTcv
     map.KtransxT.cv <- mat_data$mapKtransxTcv
     map.tlagxT.cv <- mat_data$maptlagxTcv
     map.kepxT.cv <- mat_data$mapkepxTcv
     map.vbxT.cv <- mat_data$mapvbxTcv
     map.fitfailuresxT <- mat_data$mapfitfailuresxT
     map.KtransT <- mat_data$mapKtransT
     map.kepT <- mat_data$mapkepT
     map.veT <- mat_data$mapveT
     map.fitfailuresT <- mat_data$mapfitfailuresT
     mask.roi <- mat_data$maskroi
     param.est.medianT <- mat_data$paramestmedianT
     param.est.medianxT <- mat_data$paramestmedianxT
     cc.median <- mat_data$ccmedian
     roi.median.fitted <- mat_data$roimedianfitted
     param.est.whole.roi <- mat_data$paramestwholeroi
     map.AIC.compare <- mat_data$mapAICcompare
     map.AIC.T <- mat_data$mapAICT
     map.EF <- mat_data$mapEF
     map.AIC.xT <- mat_data$mapAICxT
     roi.median.fitted.Tofts <- mat_data$roimedianfittedTofts
     param.est.whole.roi.Tofts <- mat_data$paramestwholeroiTofts
     roi.median.fitted.xTofts <- mat_data$roimedianfittedxTofts
     param.est.whole.roi.xTofts <- mat_data$paramestwholeroixTofts
     cv.whole.roi.xTofts <- mat_data$cvwholeroixTofts
     params.xTofts <- mat_data$paramsxTofts
     rm(mat_data)
     }
     if (file.format == "RData") {
     load(file)
     args <- dcemri.data$args
     results_file <- args$resultsfile
     ID.visit <- args$IDvisit
     slice <- args$slice
     method.optimization <- args$methodoptimization
     show.rt.fits <- args$showrtfits
     param.for.avdt <- param.for.avdt
     range.map <- args$rangemap
     cutoff.map <- args$cutoffmap
     lo <- args$lo
     p0.xT <- dcemri.data$p0xT
     p0.T <- dcemri.data$p0T
     AIF.shift <- args$AIFshift
     est.per.voxel.tlag <- args$estpervoxeltlag
     nx <- dcemri.data$nx
     ny <- dcemri.data$ny
     nt <- dcemri.data$nt
     export.matlab <- args$exportmatlab
     map_cc_slice <- dcemri.data$cc
     map_cc_roi <- dcemri.data$ccroi
     map.times <- dcemri.data$maptimes
     aif <- dcemri.data$aif
     aif.shifted <- dcemri.data$aifshifted
     map.KtransT.cv <- dcemri.data$mapKtransTcv
     map.kepT.cv <- dcemri.data$mapkepTcv
     map.KtransxT.cv <- dcemri.data$mapKtransxTcv
     map.tlagxT.cv <- dcemri.data$maptlagxTcv
     map.kepxT.cv <- dcemri.data$mapkepxTcv
     map.vbxT.cv <- dcemri.data$mapvbxTcv
     map.KtransxT <- dcemri.data$mapKtransxT
     map.tlagxT <- dcemri.data$maptlagxT
     map.kepxT <- dcemri.data$mapkepxT
     map.vexT <- dcemri.data$mapvexT
     map.vbxT <- dcemri.data$mapvbxT
     map.fitfailuresxT <- dcemri.data$mapfitfailuresxT
     map.KtransT <- dcemri.data$mapKtransT
     map.kepT <- dcemri.data$mapkepT
     map.veT <- dcemri.data$mapveT
     map.fitfailuresT <- dcemri.data$mapfitfailuresT
     mask.roi <- dcemri.data$maskroi
     param.est.medianT <- dcemri.data$paramestmedianT
     param.est.medianxT <- dcemri.data$paramestmedianxT
     cc.median <- dcemri.data$ccmedian
     roi.median.fitted <- dcemri.data$roimedianfitted
     param.est.whole.roi <- dcemri.data$paramestwholeroi
     map.AIC.compare <- dcemri.data$mapAICcompare
     map.AIC.T <- dcemri.data$mapAICT
     map.EF <- dcemri.data$mapEF
     map.AIC.xT <- dcemri.data$mapAICxT
     roi.median.fitted.Tofts <- dcemri.data$roimedianfittedTofts
     param.est.whole.roi.Tofts <- dcemri.data$paramestwholeroiTofts
     roi.median.fitted.xTofts <- dcemri.data$roimedianfittedxTofts
     param.est.whole.roi.xTofts <- dcemri.data$paramestwholeroixTofts
     cv.whole.roi.xTofts <- dcemri.data$cvwholeroixTofts
     params.xTofts <- dcemri.data$paramsxTofts
     rm(dcemri.data)
     }
     modeltype1 <- "xTofts"
     modeltype2 <- "Tofts"
     }
     if (file.original == TRUE) {
     cat("fitting", modeltype1, "and", modeltype2,
     "models to ROI voxels...", "\n")
     ptm <- proc.time()[3]
     t <- map.times
     map.KtransxT <- matrix(NA, nrow = nx, ncol = ny)
     map.kepxT <- matrix(NA, nrow = nx, ncol = ny)
     map.vexT <- matrix(NA, nrow = nx, ncol = ny)
     map.vbxT <- matrix(NA, nrow = nx, ncol = ny)
     map.tlagxT <- matrix(NA, nrow = nx, ncol = ny)
     map.KtransxT.cv <- matrix(NA, nrow = nx, ncol = ny)
     map.kepxT.cv <- matrix(NA, nrow = nx, ncol = ny)
     map.vbxT.cv <- matrix(NA, nrow = nx, ncol = ny)
     map.tlagxT.cv <- matrix(NA, nrow = nx, ncol = ny)
     map.fitfailuresxT <- matrix(NA, nrow = nx, ncol = ny)
     map.OptimValuexT <- matrix(NA, nrow = nx, ncol = ny)
     map.KtransT <- matrix(NA, nrow = nx, ncol = ny)
     map.kepT <- matrix(NA, nrow = nx, ncol = ny)
     map.veT <- matrix(NA, nrow = nx, ncol = ny)
     map.KtransT.cv <- matrix(NA, nrow = nx, ncol = ny)
     map.kepT.cv <- matrix(NA, nrow = nx, ncol = ny)
     map.fitfailuresT <- matrix(NA, nrow = nx, ncol = ny)
     map.OptimValueT <- matrix(NA, nrow = nx, ncol = ny)
     map.AIC.compare <- matrix(0, nrow = nx, ncol = ny)
     map.AIC.T <- matrix(NA, nrow = nx, ncol = ny)
     map.EF <- matrix(NA, nrow = nx, ncol = ny)
     map.AIC.xT <- matrix(NA, nrow = nx, ncol = ny)
     cc_fittedxT <- array(0, dim = c(nx, ny, nt))
     cc_fittedT <- array(0, dim = c(nx, ny, nt))
     nv <- 1
     nv1_q <- trunc(quantile(1:length(mask.roi[mask.roi ==
     1]), probs = c(0.1, 0.2, 0.3, 0.4, 0.5, 0.6,
     0.7, 0.8, 0.9, 1)))
     if (show.rt.fits == TRUE)
     dev.new(xpos = 3500, ypos = 0)
     ptm_slice <- proc.time()[3]
     GTzero <- function(x) {
     length(as.vector(x > 0)[as.vector(x > 0) ==
     TRUE])/length(x)
     }
     for (x in 1:nx) {
     for (y in 1:ny) {
     if (mask.roi[x, y] == 1 & GTzero(map_cc_slice[x,
     y, ]) <= fracGTzero) {
     map.fitfailuresxT[x, y] <- -2
     map.fitfailuresT[x, y] <- -2
     }
     if (mask.roi[x, y] == 1 & GTzero(map_cc_slice[x,
     y, ]) > fracGTzero)
     map.EF[x, y] <- 1
     if (mask.roi[x, y] == 1 & GTzero(map_cc_slice[x,
     y, ]) > fracGTzero) {
     nv <- nv + 1
     roi <- map_cc_slice[x, y, ]
     fix.tlag <- TRUE
     roi.model <- roi.modelxT
     if (verbose == TRUE) {
     cat("x =", x, "\n")
     cat("y =", y, "\n")
     cat("contrast agent curve =", roi, "\n")
     cat("fitting xTofts model to voxel data...")
     }
     if (method.optimization == "L-BFGS-B")
     fit_roi <- try(optim(p0.xT, Obj_roi,
     model = roi.model, t = map.times, dt = diff(map.times),
     cp = aif, roi = roi, method = "L-BFGS-B",
     lower = c(lo, lo, lo), upper = c(Inf,
     Inf, Inf), hessian = TRUE), silent = try.silent)
     if (method.optimization != "L-BFGS-B")
     fit_roi <- try(optim(p0.xT, Obj_roi,
     model = roi.model, t = map.times, dt = diff(map.times),
     cp = aif, roi = roi, method = method.optimization,
     hessian = TRUE), silent = try.silent)
     if (verbose == TRUE)
     cat("done", "\n")
     if (class(fit_roi) != "try-error") {
     if (fit_roi$convergence == 0) {
     RSS <- sum((roi.model(p = fit_roi$par,
     t = map.times, dt = diff(map.times),
     cp = aif) - roi)^2)
     param_est <- fit_roi$par
     nD <- nt
     nP <- length(param_est)
     hess <- fit_roi$hessian
     cov <- try((nP * RSS/(nD - nP)) * solve(hess),
     silent = try.silent)
     if (class(cov) != "try-error") {
     sd <- try(sqrt(diag(cov)), silent = try.silent)
     if (class(sd) != "try-error") {
     cv <- sd/param_est * 100
     cv.roi.voxel <- as.numeric(format(cv,
     digits = 1))
     map.KtransxT.cv[x, y] <- cv.roi.voxel[1]
     map.kepxT.cv[x, y] <- cv.roi.voxel[2]
     map.vbxT.cv[x, y] <- cv.roi.voxel[3]
     if (est.per.voxel.tlag == TRUE)
     map.tlagxT.cv[x, y] <- cv.roi.voxel[4]
     }
     }
     map.KtransxT[x, y] <- fit_roi$par[1]
     map.kepxT[x, y] <- fit_roi$par[2]
     if (map.kepxT[x, y] < 1e-05)
     map.kepxT[x, y] <- 1e-05
     map.vexT[x, y] <- fit_roi$par[1]/map.kepxT[x,
     y]
     map.vbxT[x, y] <- fit_roi$par[3]
     if (est.per.voxel.tlag == TRUE)
     map.tlagxT[x, y] <- fit_roi$par[4]
     map.OptimValuexT[x, y] <- fit_roi$value
     }
     }
     if (class(fit_roi) == "try-error") {
     if (verbose == TRUE)
     cat("...but a <try-error> occurred",
     "\n")
     map.fitfailuresxT[x, y] <- 99
     }
     if (class(fit_roi) != "try-error")
     map.fitfailuresxT[x, y] <- fit_roi$convergence
     if (class(fit_roi) == "try-error") {
     map.KtransxT[x, y] <- NA
     map.kepxT[x, y] <- NA
     map.vexT[x, y] <- NA
     map.vbxT[x, y] <- NA
     map.tlagxT[x, y] <- NA
     }
     if (class(fit_roi) != "try-error") {
     if (fit_roi$convergence > 0) {
     map.KtransxT[x, y] <- NA
     map.kepxT[x, y] <- NA
     map.vexT[x, y] <- NA
     map.vbxT[x, y] <- NA
     map.tlagxT[x, y] <- NA
     }
     }
     if (verbose == TRUE)
     cat("simulating xTofts model at estimated parameter values...")
     if (class(fit_roi) != "try-error") {
     if (fit_roi$convergence == 0) {
     simulation <- roi.model(p = fit_roi$par,
     t = map.times, dt = diff(map.times),
     cp = aif)
     cc_fittedxT[x, y, ] <- simulation
     }
     }
     if (verbose == TRUE)
     cat("done", "\n")
     roi.model <- roi.modelT
     if (verbose == TRUE)
     cat("fitting Tofts model to voxel data...")
     if (method.optimization == "L-BFGS-B")
     fit_roiT <- try(optim(p0.T, Obj_roi,
     model = roi.model, t = map.times, dt = diff(map.times),
     cp = aif, roi = roi, method = "L-BFGS-B",
     lower = lower.wholeT, upper = upper.wholeT,
     hessian = TRUE), silent = try.silent)
     if (method.optimization != "L-BFGS-B")
     fit_roiT <- try(optim(p0.T, Obj_roi,
     model = roi.model, t = map.times, dt = diff(map.times),
     cp = aif, roi = roi, method = method.optimization,
     hessian = TRUE), silent = try.silent)
     if (verbose == TRUE)
     cat("done", "\n")
     if (class(fit_roiT) != "try-error") {
     if (fit_roiT$convergence == 0) {
     RSS <- sum((roi.model(p = fit_roiT$par,
     t = map.times, dt = diff(map.times),
     cp = aif) - roi)^2)
     nD <- nt
     nP <- length(param_est)
     param_est <- fit_roiT$par
     df <- nt - length(param_est)
     hess <- fit_roiT$hessian
     cov <- try((nP * RSS/(nD - nP)) * solve(hess),
     silent = try.silent)
     if (class(cov) != "try-error") {
     sd <- try(sqrt(diag(cov)), silent = try.silent)
     if (class(sd) != "try-error") {
     cv <- sd/param_est * 100
     cv.roi.voxel <- as.numeric(format(cv,
     digits = 1))
     map.KtransT.cv[x, y] <- cv.roi.voxel[1]
     map.kepT.cv[x, y] <- cv.roi.voxel[2]
     }
     }
     map.KtransT[x, y] <- fit_roiT$par[1]
     map.kepT[x, y] <- fit_roiT$par[2]
     if (map.kepT[x, y] < 1e-05)
     map.kepT[x, y] <- 1e-05
     map.veT[x, y] <- fit_roiT$par[1]/map.kepT[x,
     y]
     map.OptimValueT[x, y] <- fit_roiT$value
     }
     }
     if (class(fit_roiT) == "try-error") {
     map.fitfailuresT[x, y] <- 99
     if (verbose == TRUE)
     cat("...but a <try-error> occurred",
     "\n")
     }
     if (class(fit_roiT) == "try-error") {
     map.KtransT[x, y] <- NA
     map.kepT[x, y] <- NA
     map.veT[x, y] <- NA
     }
     if (class(fit_roiT) != "try-error") {
     if (fit_roiT$convergence > 0) {
     map.KtransT[x, y] <- NA
     map.kepT[x, y] <- NA
     map.veT[x, y] <- NA
     }
     }
     if (class(fit_roiT) != "try-error")
     map.fitfailuresT[x, y] <- fit_roiT$convergence
     if (verbose == TRUE)
     cat("simulating Tofts model at estimated parameter values...")
     if (class(fit_roiT) != "try-error") {
     if (fit_roiT$convergence == 0) {
     simulation_2 <- roi.model(p = fit_roiT$par,
     t = map.times, dt = diff(map.times),
     cp = aif)
     cc_fittedT[x, y, ] <- simulation_2
     }
     }
     if (verbose == TRUE)
     cat("done", "\n")
     roi.model <- roi.modelxT
     if (class(fit_roi) != "try-error" & class(fit_roiT) !=
     "try-error") {
     if (fit_roi$convergence == 0 & fit_roiT$convergence ==
     0) {
     if (verbose == TRUE)
     cat("calculating AICc values...")
     simulation_AIC <- simulation
     simulation_AIC_2 <- simulation_2
     np_1 <- length(fit_roiT$par)
     np_2 <- length(fit_roi$par) + 1
     AIC1 <- nt * log(fit_roiT$value) +
     2 * (np_1 + 1) + (2 * (np_1 + 1) *
     (np_1 + 2))/(nt - np_1 - 2)
     AIC2 <- nt * log(fit_roi$value) + 2 *
     (np_2 + 1) + (2 * (np_2 + 1) * (np_2 +
     2))/(nt - np_2 - 2)
     AIC1 <- as.numeric(format(AIC1, digits = 1))
     AIC2 <- as.numeric(format(AIC2, digits = 1))
     if (AIC2 < AIC1)
     map.AIC.compare[x, y] <- 1
     if (AIC2 >= AIC1)
     map.AIC.compare[x, y] <- 2
     map.AIC.T[x, y] <- AIC1
     map.AIC.xT[x, y] <- AIC2
     }
     }
     if (class(fit_roi) != "try-error") {
     if (fit_roi$convergence > 0) {
     map.AIC.compare[x, y] <- NA
     }
     }
     if (class(fit_roiT) != "try-error") {
     if (fit_roiT$convergence > 0) {
     map.AIC.compare[x, y] <- NA
     }
     }
     if (class(fit_roi) == "try-error" | class(fit_roiT) ==
     "try-error")
     map.AIC.compare[x, y] <- NA
     if (verbose == TRUE) {
     cat("done", "\n")
     cat("======================", "\n")
     }
     if (show.rt.fits == TRUE) {
     if (class(fit_roiT) != "try-error" &
     class(fit_roi) != "try-error") {
     if (fit_roiT$convergence == 0 & fit_roi$convergence ==
     0) {
     plot(map.times, roi, ylab = "contrast agent",
     xlab = "min", main = paste(modeltype1,
     "(red) and", modeltype2, "(blue)"),
     cex = 3)
     lines(map.times, simulation, col = "red",
     lwd = 5)
     lines(map.times, simulation_2, col = "blue",
     lwd = 2)
     }
     }
     }
     if (nv == 2)
     cat("progress: ")
     if (nv == nv1_q[[1]])
     cat("10%..")
     if (nv == nv1_q[[2]])
     cat("20%..")
     if (nv == nv1_q[[3]])
     cat("30%..")
     if (nv == nv1_q[[4]])
     cat("40%..")
     if (nv == nv1_q[[5]])
     cat("50%..")
     if (nv == nv1_q[[6]])
     cat("60%..")
     if (nv == nv1_q[[7]])
     cat("70%..")
     if (nv == nv1_q[[8]])
     cat("80%..")
     if (nv == nv1_q[[9]])
     cat("90%..", "\n")
     if (nv == nv1_q[[10]] - 10)
     cat("..10")
     if (nv == nv1_q[[10]] - 9)
     cat("..9..")
     if (nv == nv1_q[[10]] - 8)
     cat("8..")
     if (nv == nv1_q[[10]] - 7)
     cat("7..")
     if (nv == nv1_q[[10]] - 6)
     cat("6..")
     if (nv == nv1_q[[10]] - 5)
     cat("5..")
     if (nv == nv1_q[[10]] - 4)
     cat("4..")
     if (nv == nv1_q[[10]] - 3)
     cat("3..")
     if (nv == nv1_q[[10]] - 2)
     cat("2..")
     if (nv == nv1_q[[10]] - 1)
     cat("1..", "\n")
     }
     }
     }
     cat("..done in", format((proc.time()[3] - ptm)/60,
     digits = 2), "minutes.", "\n")
     cat("--------", "\n")
     graphics.off()
     KtransxT.median <- median(map.KtransxT[map.KtransxT >
     0 & map.fitfailuresxT == 0], na.rm = TRUE)
     kepxT.median <- median(map.kepxT[map.kepxT >
     0 & map.fitfailuresxT == 0], na.rm = TRUE)
     vexT.median <- median(map.vexT[map.vexT > 0 &
     map.fitfailuresxT == 0], na.rm = TRUE)
     vbxT.median <- median(map.vbxT[map.vbxT >= 0 &
     map.fitfailuresxT == 0], na.rm = TRUE)
     if (est.per.voxel.tlag == TRUE)
     tlagxT.median <- median(map.tlagxT[map.tlagxT >=
     0 & map.fitfailuresxT == 0], na.rm = TRUE)
     if (est.per.voxel.tlag == FALSE)
     tlagxT.median <- NA
     fitfailuresxT.total <- length(map.fitfailuresxT[map.fitfailuresxT >=
     1])/length(map.fitfailuresxT[map.fitfailuresxT >=
     0]) * 100
     param.est.medianxT <- list(KtransxT.median, kepxT.median,
     vexT.median, vbxT.median, tlagxT.median, fitfailuresxT.total)
     names(param.est.medianxT) <- c("Ktrans.median",
     "kep.median", "ve.median", "vb.median", "tlag.median",
     "percent.fitfailures")
     KtransT.median <- median(map.KtransT[map.KtransT >
     0 & map.fitfailuresT == 0], na.rm = TRUE)
     kepT.median <- median(map.kepT[map.kepT > 0 &
     map.fitfailuresT == 0], na.rm = TRUE)
     veT.median <- median(map.veT[map.veT > 0 & map.fitfailuresT ==
     0], na.rm = TRUE)
     fitfailuresT.total <- length(map.fitfailuresT[map.fitfailuresT >=
     1])/length(map.fitfailuresT[map.fitfailuresT >=
     0]) * 100
     param.est.medianT <- list(KtransT.median, kepT.median,
     veT.median, fitfailuresT.total)
     names(param.est.medianT) <- c("Ktrans.median",
     "kep.median", "ve.median", "percent.fitfailures")
     }
     if (file.original == TRUE) {
     if (file == "concatenate.KAT.with.KAT.checkData.RData")
     ID.visit <- paste(strsplit(results_file, split = ".mat")[[1]],
     "_s", slice, sep = "")
     if (file != "concatenate.KAT.with.KAT.checkData.RData")
     ID.visit <- paste(strsplit(file, split = ".mat")[[1]],
     "_s", slice, sep = "")
     }
     if (file.original == FALSE)
     ID.visit <- strsplit(file, split = ".mat")[[1]]
     ID.visit <- strsplit(ID.visit, split = "/")
     ID.visit <- ID.visit[[1]][length(ID.visit[[1]])]
     IDvp <- strsplit(ID.visit, split = "_")
     ID.visit.forplot <- paste(IDvp[[1]][1], ".", IDvp[[1]][2],
     ".", IDvp[[1]][3], ".", IDvp[[1]][4], sep = "")
     DATE <- date()
     if (length(strsplit(DATE, split = " ")[[1]]) ==
     2) {
     full_date <- strsplit(DATE, split = " ")[[1]]
     full_date_1 <- full_date[1]
     full_date_1 <- strsplit(full_date_1, split = " ")[[1]]
     full_date_2 <- full_date[2]
     full_date_2 <- strsplit(full_date_2, split = " ")[[1]]
     month <- full_date_1[2]
     day <- full_date_2[1]
     time <- full_date_2[2]
     year <- full_date_2[3]
     }
     if (length(strsplit(DATE, split = " ")[[1]]) ==
     1) {
     full_date <- strsplit(DATE, split = " ")[[1]]
     month <- full_date[2]
     day <- full_date[3]
     time <- full_date[4]
     year <- full_date[5]
     }
     year <- strsplit(year, split = "")[[1]]
     year <- paste(year[3], year[4], sep = "")
     time_concat <- strsplit(time, split = ":")[[1]]
     time_concat <- paste(paste(time_concat[1], time_concat[2],
     sep = ""), time_concat[3], sep = "")
     DATE <- paste(paste(paste(paste(day, month, sep = ""),
     year, sep = ""), "-", sep = ""), time_concat,
     sep = "")
     filename3 <- paste(ID.visit, "_KAT_", DATE, ".mat",
     sep = "")
     filename3 <- sub(".RData", "", filename3)
     if (file.original == FALSE) {
     roi.model <- roi.modelxT
     }
     if (param.for.avdt == "Ktrans")
     MAP <- map.KtransxT
     if (param.for.avdt == "kep")
     MAP <- map.kepxT
     if (param.for.avdt == "ve")
     MAP <- map.vexT
     if (param.for.avdt == "vb")
     MAP <- map.vbxT
     if (param.for.avdt == "tlag")
     MAP <- map.tlagxT
     x_min <- 0
     x_max <- nx
     y_min <- 0
     y_max <- ny
     for (xx in 1:nx) {
     if (sum(MAP[xx, ], na.rm = TRUE) > 0) {
     x_min <- xx - 3
     break
     }
     if (sum(MAP[xx, ], na.rm = TRUE) == 0)
     xx <- xx + 1
     }
     for (y in 1:ny) {
     if (sum(MAP[, y], na.rm = TRUE) > 0) {
     y_min <- y - 3
     break
     }
     if (sum(MAP[, y], na.rm = TRUE) == 0)
     y <- y + 1
     }
     for (xx in nx:0) {
     if (sum(MAP[xx, ], na.rm = TRUE) > 0) {
     x_max <- xx + 3
     break
     }
     if (sum(MAP[xx, ], na.rm = TRUE) == 0)
     xx <- xx - 1
     }
     for (y in ny:0) {
     if (sum(MAP[, y], na.rm = TRUE) > 0) {
     y_max <- y + 3
     break
     }
     if (sum(MAP[, y], na.rm = TRUE) == 0)
     y <- y - 1
     }
     MAP_ul_s1 <- MAP[MAP > 0]
     MAP_ul_s1 <- sort(MAP_ul_s1)
     MAP_ul <- range.map * (max(MAP_ul_s1[1:length(MAP_ul_s1) *
     cutoff.map]))
     if (file.original == FALSE) {
     if (param.for.avdt == "Ktrans")
     MAP <- map.KtransxT
     if (param.for.avdt == "kep")
     MAP <- map.kepxT
     if (param.for.avdt == "ve")
     MAP <- map.vexT
     if (param.for.avdt == "vb")
     MAP <- map.vbxT
     if (param.for.avdt == "tlag")
     MAP <- map.tlagxT
     if (file.format == "matlab") {
     if (param.for.avdt == "Ktrans")
     param.median <- param.est.medianxT[, , 1]$Ktrans.median[1]
     if (param.for.avdt == "kep")
     param.median <- param.est.medianxT[, , 1]$kep.median[1]
     if (param.for.avdt == "ve")
     param.median <- param.est.medianxT[, , 1]$ve.median[1]
     if (param.for.avdt == "vb")
     param.median <- param.est.medianxT[, , 1]$vb.median[1]
     if (param.for.avdt == "tlag")
     param.median <- param.est.medianxT[, , 1]$tlag.median[1]
     }
     if (file.format == "RData") {
     if (param.for.avdt == "Ktrans")
     param.median <- param.est.medianxT$Ktrans.median
     if (param.for.avdt == "kep")
     param.median <- param.est.medianxT$kep.median
     if (param.for.avdt == "ve")
     param.median <- param.est.medianxT$ve.median
     if (param.for.avdt == "vb")
     param.median <- param.est.medianxT$vb.median
     if (param.for.avdt == "tlag")
     param.median <- param.est.medianxT$tlag.median
     }
     MAP_for_plot <- MAP
     MAP_for_plot[MAP_for_plot < 0] <- 0
     MAP_for_plot[MAP_for_plot >= MAP_ul * 0.99] <- MAP_ul *
     0.99
     dev.new(width = 6, height = 6, xpos = 1860, ypos = 578)
     plot(map.times, cc.median, xlab = "min", ylab = "contrast agent",
     main = paste(modeltype1, "(red) and", modeltype2,
     "(blue) fitted to median whole ROI data"),
     cex.main = 1, cex.axis = 1, cex.lab = 1, cex = 2)
     lines(map.times, roi.median.fitted.xTofts, col = "red",
     lwd = 5)
     lines(map.times, roi.median.fitted.Tofts, col = "blue",
     lwd = 2)
     text(x = 0.6 * max(map.times), y = 0.3 * max(cc.median,
     na.rm = TRUE), labels = "EXTENDED TOFTS PARAMS",
     cex = 1)
     text(x = 0.6 * max(map.times), y = 0.24 * max(cc.median,
     na.rm = TRUE), labels = paste(expression(K^trans),
     " = ", format(param.est.whole.roi.xTofts$Ktrans,
     digits = 3), " 1/min (", cv.whole.roi.xTofts[1],
     "%)", sep = ""), cex = 1)
     text(x = 0.6 * max(map.times), y = 0.18 * max(cc.median,
     na.rm = TRUE), labels = paste(expression(k_ep),
     " = ", format(param.est.whole.roi.xTofts$kep,
     digits = 3), " 1/min (", cv.whole.roi.xTofts[2],
     "%)", sep = ""), cex = 1)
     text(x = 0.6 * max(map.times), y = 0.12 * max(cc.median,
     na.rm = TRUE), labels = paste(expression(v_b),
     " = ", format(param.est.whole.roi.xTofts$vb,
     digits = 3), " dimensionless (", cv.whole.roi.xTofts[3],
     "%)", sep = ""), cex = 1)
     if (AIF.shift != "NONE")
     text(x = 0.6 * max(map.times), y = 0.06 * max(cc.median,
     na.rm = TRUE), labels = paste(expression(t_lag),
     " = ", format(60 * param.est.whole.roi.xTofts$tlag,
     digits = 3), " sec (", cv.whole.roi.xTofts[4],
     "%)", sep = ""), cex = 1)
     dev.new(width = 6, height = 6, xpos = 1860, ypos = 0)
     image(map.AIC.compare, xlim = c(x_min/nx, x_max/nx),
     ylim = c(y_min/ny, y_max/ny), col = palette(colorRampPalette(c("black",
     "red", "blue"))(3)), main = paste("best fit:",
     modeltype1, "(red) and", modeltype2, "(blue)"))
     image(map.AIC.compare, xlim = c(x_min/nx, x_max/nx),
     ylim = c(y_min/ny, y_max/ny), col = palette(colorRampPalette(c("black",
     "red", "blue"))(3)), main = paste("best fit:",
     modeltype1, "(red) and", modeltype2, "(blue)"))
     dev.new(width = 12.75, height = 12.75, xpos = 238,
     ypos = 0)
     image(MAP_for_plot, col = palette(colorRampPalette(c("black",
     "red", "yellow"))(1000)), xlim = c(x_min/nx,
     x_max/nx), ylim = c(y_min/ny, y_max/ny), zlim = c(0,
     MAP_ul), main = paste("Model Type=", modeltype1,
     " Median ", param.for.avdt, "=", format(param.median,
     digit = 2), " VisitID=", strsplit(args$IDvisit,
     split = "_")[[1]][1], " slice=", args$slice,
     sep = ""))
     image(MAP_for_plot, col = palette(colorRampPalette(c("black",
     "red", "yellow"))(1000)), xlim = c(x_min/nx,
     x_max/nx), ylim = c(y_min/ny, y_max/ny), zlim = c(0,
     MAP_ul), main = paste("Model Type=", modeltype1,
     " Median ", param.for.avdt, "=", format(param.median,
     digit = 2), " VisitID=", strsplit(args$IDvisit,
     split = "_")[[1]][1], " slice=", args$slice,
     sep = ""))
     text(x_min/nx + (x_max/nx - x_min/nx) * 0.97,
     y_min/ny + (y_max/ny - y_min/ny) * 0.99, "close",
     col = "green")
     text(x_min/nx + (x_max/nx - x_min/nx) * 0.05,
     y_min/ny + (y_max/ny - y_min/ny) * 0.99, "print to PDF",
     col = "green")
     text(x_min/nx + (x_max/nx - x_min/nx) * 0.84,
     y_min/ny + (y_max/ny - y_min/ny) * 0.01, paste("KAT for DCEMRI v",
     KAT.version, ", Genentech PTPK", sep = ""),
     col = "darkgrey")
     legend <- seq(0, MAP_ul, by = 0.001)
     dim(legend) <- c(1, length(legend))
     dev.new(width = 2.5, height = 12.75, xpos = 0,
     ypos = 0)
     image(legend, col = palette(colorRampPalette(c("black",
     "red", "yellow"))(1000)), zlim = c(0, MAP_ul),
     xaxt = "n", yaxt = "n", xlab = "0", main = format(MAP_ul,
     digits = 3), ylab = expression(min^-1), cex.lab = 1.25,
     cex.main = 1.05)
     image(legend, col = palette(colorRampPalette(c("black",
     "red", "yellow"))(1000)), zlim = c(0, MAP_ul),
     xaxt = "n", yaxt = "n", xlab = "0", main = format(MAP_ul,
     digits = 3), ylab = expression(min^-1), cex.lab = 1.25,
     cex.main = 1.05)
     if (AIF.shift == "ARTERY") {
     dev.new(width = 5.15, height = 4.4, xpos = 1500,
     ypos = 0)
     plot(map.times, aif, ylim = c(0, max(aif, aif.shifted)),
     xlab = "min", ylab = "contrast agent", main = paste("Shifted Arterial Input Function (",
     format(param.est.whole.roi.xTofts$tlag *
     60, digits = 3), " sec)", sep = ""),
     type = "n")
     lines(map.times, aif, col = "red")
     lines(map.times, aif.shifted)
     legend(x = 0.55 * max(map.times), y = max(aif,
     aif.shifted), c("raw AIF", "shifted AIF"),
     c("red", "black"))
     }
     if (AIF.shift == "VEIN") {
     dev.new(width = 5.15, height = 4.4, xpos = 1500,
     ypos = 0)
     plot(map.times, aif, ylim = c(0, max(aif, aif.shifted)),
     xlab = "min", ylab = "contrast agent", main = paste("Shifted Venous Input Function (",
     format(param.est.whole.roi.xTofts$tlag *
     60, digits = 3), " sec)", sep = ""),
     type = "n")
     lines(map.times, aif, col = "blue")
     lines(map.times, aif.shifted)
     legend(x = 0.55 * max(map.times), y = max(aif,
     aif.shifted), c("raw VIF", "shifted VIF"),
     c("blue", "black"))
     }
     if (AIF.shift == "NONE") {
     dev.new(width = 5.15, height = 4.4, xpos = 1500,
     ypos = 0)
     plot(map.times, aif, ylim = c(0, max(aif)),
     xlab = "min", ylab = "contrast agent", main = "Vascular Input Function",
     type = "n")
     lines(map.times, aif)
     legend(x = 0.55 * max(map.times), y = max(aif),
     c("raw VIF"), c("black"))
     }
     dev.set(4)
     inf <- 1
     newplot <- 1
     newplot2 <- 1
     legend_count <- 1
     legend_labels <- 1:1000
     legend_matrix <- matrix(0, ncol = ny, nrow = nx)
     cat("---", "\n")
     while (inf == 1) {
     z <- locator(1, type = "o", col = "green")
     xx <- round(z$x * (nx - 1) + 1)
     yy <- round(z$y * (ny - 1) + 1)
     if (legend_matrix[xx, yy] == 0) {
     legend(z$x, z$y, legend_labels[legend_count],
     col = "green", text.col = "green")
     legend_matrix[xx, yy] <- 1
     cat("Voxel Number/Coordinates: n=", legend_count,
     ", x=", xx, ", y=", yy, "\n", sep = "")
     cat("Parameter estimates (xTofts): Ktrans=",
     format(map.KtransxT[xx, yy], digits = 3),
     ", ve=", format(map.KtransxT[xx, yy]/map.kepxT[xx,
     yy], digits = 3), ", vb=", format(map.vbxT[xx,
     yy], digits = 3), ", tlag=", format(map.tlagxT[xx,
     yy], digits = 3), "\n", sep = "")
     cat("%CVs of Parameter estimates (xTofts): Ktrans=",
     format(map.KtransxT.cv[xx, yy], digits = 2),
     ", kep (Ktrans/ve)=", format(map.kepxT.cv[xx,
     yy], digits = 2), ", vb=", format(map.vbxT.cv[xx,
     yy], digits = 2), ", tlag=", format(map.tlagxT.cv[xx,
     yy], digits = 2), "\n", sep = "")
     cat("Parameter estimates (Tofts): Ktrans=",
     format(map.KtransT[xx, yy], digits = 3),
     ", ve=", format(map.KtransT[xx, yy]/map.kepT[xx,
     yy], digits = 3), "\n", sep = "")
     cat("%CVs of Parameter estimates (Tofts): Ktrans=",
     format(map.KtransT.cv[xx, yy], digits = 2),
     ", kep (Ktrans/ve)=", format(map.kepT.cv[xx,
     yy], digits = 3), "\n", sep = "")
     cat("---", "\n")
     legend_count <- legend_count + 1
     }
     xdim <- x_max - x_min
     ydim <- y_max - y_min
     if (xx > (x_max - 0.1 * xdim) && yy > (y_max -
     0.02 * ydim)) {
     graphics.off()
     dev.off()
     }
     xx_old <- xx
     yy_old <- yy
     conc <- 1:nt
     for (i in 1:nt) conc[i] <- map_cc_slice[xx,
     yy, i]
     if (xx < (x_min + 0.12 * xdim) && yy > (y_max -
     0.02 * ydim)) {
     pdf(file = paste(results_file, "-SUMMARY.pdf",
     sep = ""), height = 12, width = 15)
     layout(matrix(c(1, 2, 3, 4, 5, 6), 2, 3,
     byrow = TRUE), widths = c(1.5, 5.5, 5.5))
     par(omi = c(0.15, 0.15, 0.15, 0.15))
     image(legend, col = palette(colorRampPalette(c("black",
     "red", "yellow"))(1000)), zlim = c(0, MAP_ul),
     xaxt = "n", yaxt = "n", xlab = "0", main = format(MAP_ul,
     digits = 3), ylab = expression(min^-1),
     cex.lab = 1.5, cex.main = 1.5, cex.axis = 1.5)
     image(legend, col = palette(colorRampPalette(c("black",
     "red", "yellow"))(1000)), zlim = c(0, MAP_ul),
     xaxt = "n", yaxt = "n", xlab = "0", main = format(MAP_ul,
     digits = 3), ylab = expression(min^-1),
     cex.lab = 1.5, cex.main = 1.5, cex.axis = 1.5,
     add = TRUE)
     image(MAP_for_plot, col = palette(colorRampPalette(c("black",
     "red", "yellow"))(1000)), xlim = c(x_min/nx,
     x_max/nx), ylim = c(y_min/ny, y_max/ny),
     zlim = c(0, MAP_ul), main = paste("Model Type=",
     modeltype1, " Median ", param.for.avdt,
     "=", format(param.median, digit = 2),
     " VisitID=", strsplit(args$IDvisit,
     split = "_")[[1]][1], " slice=",
     args$slice, sep = ""), cex.axis = 1.5)
     image(MAP_for_plot, col = palette(colorRampPalette(c("black",
     "red", "yellow"))(1000)), xlim = c(x_min/nx,
     x_max/nx), ylim = c(y_min/ny, y_max/ny),
     zlim = c(0, MAP_ul), main = paste("Model Type=",
     modeltype1, " Median ", param.for.avdt,
     "=", format(param.median, digit = 2),
     " VisitID=", strsplit(args$IDvisit,
     split = "_")[[1]][1], " slice=",
     args$slice, sep = ""), cex.axis = 1.5,
     add = TRUE)
     plot(map.times, cc.median, xlab = "min",
     ylab = "contrast agent", main = "median contrast agent conc; Tofts=blue, xTofts=red",
     cex = 3, cex.axis = 1.5, cex.lab = 1.5)
     text(x = 0.6 * max(map.times), y = 0.25 *
     max(cc.median, na.rm = TRUE), labels = "EXTENDED TOFTS PARAMS",
     cex = 1.5)
     text(x = 0.6 * max(map.times), y = 0.2 *
     max(cc.median, na.rm = TRUE), labels = paste(expression(K^trans),
     " = ", format(param.est.whole.roi.xTofts$Ktrans,
     digits = 3), " 1/min (", cv.whole.roi.xTofts[1],
     "%)", sep = ""), cex = 1.5)
     text(x = 0.6 * max(map.times), y = 0.15 *
     max(cc.median, na.rm = TRUE), labels = paste(expression(v_e),
     " = ", format(param.est.whole.roi.xTofts$ve,
     digits = 3), " 1/min (", cv.whole.roi.xTofts[2],
     "%)", sep = ""), cex = 1.5)
     text(x = 0.6 * max(map.times), y = 0.1 *
     max(cc.median, na.rm = TRUE), labels = paste(expression(v_b),
     " = ", format(param.est.whole.roi.xTofts$vb,
     digits = 3), " dimensionless (", cv.whole.roi.xTofts[3],
     "%)", sep = ""), cex = 1.5)
     if (AIF.shift != "NONE")
     text(x = 0.6 * max(map.times), y = 0.05 *
     max(cc.median, na.rm = TRUE), labels = paste(expression(t_lag),
     " = ", format(60 * param.est.whole.roi.xTofts$tlag,
     digits = 3), " sec (", cv.whole.roi.xTofts[4],
     "%)", sep = ""), cex = 1.5)
     lines(map.times, roi.median.fitted.xTofts,
     col = "red", lwd = 5)
     lines(map.times, roi.median.fitted.Tofts,
     col = "blue", lwd = 2)
     text(x = 0.5 * max(map.times), 1 * max(cc.median,
     na.rm = TRUE), paste("R package version =",
     KAT.version), cex = 1.5)
     image(array(1:2, dim = c(1, 2)), col = palette(colorRampPalette(c("red",
     "blue"))(2)), zlim = c(0, 2), xaxt = "n",
     yaxt = "n", xlab = modeltype1, main = modeltype2,
     cex.lab = 1.5, cex.main = 1.5, cex.axis = 1.5)
     image(array(1:2, dim = c(1, 2)), col = palette(colorRampPalette(c("red",
     "blue"))(2)), zlim = c(0, 2), xaxt = "n",
     yaxt = "n", xlab = modeltype1, main = modeltype2,
     cex.lab = 1.5, cex.main = 1.5, cex.axis = 1.5,
     add = TRUE)
     image(map.AIC.compare, xlim = c(x_min/nx,
     x_max/nx), ylim = c(y_min/ny, y_max/ny),
     col = palette(colorRampPalette(c("black",
     "red", "blue"))(3)), main = "Summary of model discrimination analysis",
     cex.axis = 1.5, cex.lab = 1.5)
     image(map.AIC.compare, xlim = c(x_min/nx,
     x_max/nx), ylim = c(y_min/ny, y_max/ny),
     col = palette(colorRampPalette(c("black",
     "red", "blue"))(3)), main = "Summary of model discrimination analysis",
     cex.axis = 1.5, cex.lab = 1.5, add = TRUE)
     if (AIF.shift == "ARTERY") {
     plot(map.times, aif, ylim = c(0, max(aif,
     aif.shifted)), xlab = "min", ylab = "contrast agent",
     main = paste("Shifted Arterial Input Function (",
     format(param.est.whole.roi.xTofts$tlag *
     60, digits = 3), " sec)", sep = ""),
     type = "n", cex = 3, cex.axis = 1.5,
     cex.lab = 1.5)
     lines(map.times, aif, col = "red")
     lines(map.times, aif.shifted)
     legend(x = 0.55 * max(map.times), y = max(aif,
     aif.shifted), c("raw AIF", "shifted AIF"),
     c("red", "black"), cex = 2)
     }
     if (AIF.shift == "VEIN") {
     plot(map.times, aif, ylim = c(0, max(aif,
     aif.shifted)), xlab = "min", ylab = "contrast agent",
     main = paste("Shifted Venous Input Function (",
     format(param.est.whole.roi.xTofts$tlag *
     60, digits = 3), " sec)", sep = ""),
     type = "n", cex = 3, cex.axis = 1.5,
     cex.lab = 1.5)
     lines(map.times, aif, col = "blue")
     lines(map.times, aif.shifted)
     legend(x = 0.55 * max(map.times), y = max(aif,
     aif.shifted), c("raw VIF", "shifted VIF"),
     c("blue", "black"), cex = 2)
     }
     if (AIF.shift == "NONE") {
     plot(map.times, aif, ylim = c(0, max(aif)),
     xlab = "min", ylab = "contrast agent",
     main = "Vascular Input Function", type = "n",
     cex = 3, cex.axis = 1.5, cex.lab = 1.5)
     lines(map.times, aif)
     legend(x = 0.55 * max(map.times), y = max(aif),
     c("raw VIF"), c("black"), cex = 2)
     }
     dev.off()
     cat("image printed to pdf.", "\n")
     cat("---", "\n")
     }
     if (newplot == 1) {
     dev.new(width = 5.15, height = 4.4, xpos = 1500,
     ypos = 432)
     newplot <- 2
     }
     dev.set(7)
     plot(map.times, conc, xlab = "min", ylab = "contrast agent",
     ylim = c(-max(conc, na.rm = TRUE)/5, 1.4 *
     max(conc, na.rm = TRUE)), cex = 1.5, main = paste(paste("red=",
     modeltype1, ", blue=", modeltype2, " (",
     sep = ""), paste("x=", round(xx_old), ", y=",
     round(yy_old), ")", sep = ""), sep = ""))
     value_xTofts <- MAP[xx, yy]
     if (is.finite(value_xTofts) == FALSE)
     value_xTofts <- 0
     if (value_xTofts != 0) {
     if (est.per.voxel.tlag == TRUE)
     paramsxT <- c(map.KtransxT[xx, yy], map.kepxT[xx,
     yy], map.vbxT[xx, yy], map.tlagxT[xx,
     yy])
     if (est.per.voxel.tlag == FALSE)
     paramsxT <- c(map.KtransxT[xx, yy], map.kepxT[xx,
     yy], map.vbxT[xx, yy], param.est.whole.roi.xTofts$tlag)
     paramsT <- c(map.KtransT[xx, yy], map.kepT[xx,
     yy])
     roi.model <- roi.modelxT
     simulation <- roi.model(p = paramsxT, t = map.times,
     dt = diff(map.times), cp = aif)
     roi.model <- roi.modelT
     simulation_2 <- roi.model(p = paramsT, t = map.times,
     dt = diff(map.times), cp = aif)
     lines(map.times, simulation_2, col = "blue",
     lwd = 2)
     lines(map.times, simulation, col = "red",
     lwd = 2, lty = 2)
     roi.model <- roi.modelxT
     }
     text(max(map.times)/4.5, 1.4 * max(conc, na.rm = TRUE),
     "Fitted xTofts params")
     text(max(map.times)/4.5, 1.4 * max(conc, na.rm = TRUE) *
     0.93, paste(paste("Ktrans =", format(map.KtransxT[xx,
     yy], digits = 3)), "min^-1"))
     text(max(map.times)/4.5, 1.4 * max(conc, na.rm = TRUE) *
     0.86, paste("ve =", format(map.vexT[xx, yy],
     digits = 3)))
     text(max(map.times)/4.5, 1.4 * max(conc, na.rm = TRUE) *
     0.79, paste("vb =", format(map.vbxT[xx, yy],
     digits = 3)))
     if (est.per.voxel.tlag == TRUE)
     text(max(map.times)/4.5, 1.4 * max(conc,
     na.rm = TRUE) * 0.72, paste("tlag =", format(map.tlagxT[xx,
     yy], digits = 3)))
     if (newplot2 == 1) {
     dev.new(width = 5.15, height = 3, xpos = 1500,
     ypos = 864)
     newplot2 <- 2
     }
     else dev.set(8)
     if (value_xTofts != 0) {
     simulation_AIC <- simulation
     simulation_AIC_2 <- simulation_2
     if (length(simulation) == length(conc)) {
     plot(map.times, conc - simulation, xlab = "min",
     ylab = "predicated - measured", cex = 1.5,
     col = "red", main = paste("AIC(Tofts)=",
     map.AIC.T[xx, yy], " AIC(xTofts)=",
     map.AIC.xT[xx, yy], sep = ""))
     lines(locfit(conc - simulation ~ map.times,
     acri = "ici"), col = "red", lwd = 2)
     abline(h = 0, col = "black", lwd = 2)
     if (is.finite(simulation_2[1]) == TRUE) {
     points(map.times, conc - simulation_2,
     xlab = "min", cex = 1.5, col = "blue")
     lines(locfit(conc - simulation_2 ~ map.times,
     acri = "ici"), col = "blue", lwd = 2,
     lty = 2)
     }
     }
     }
     dev.set(4)
     }
     }
     if (file.original == TRUE) {
     ptm <- proc.time()[3]
     proc.time.total <- format((proc.time()[3] - ptm_total)/60,
     digits = 2)
     args <- list(as.character(file), as.character(results_file),
     as.character(method.optimization), show.rt.fits,
     as.character(param.for.avdt), range.map, cutoff.map,
     export.matlab, export.RData, verbose, show.errors,
     try.silent, fracGTzero, AIF.shift, slice, ID.visit,
     est.per.voxel.tlag)
     names(args) <- c("file", "resultsfile", "methodoptimization",
     "showrtfits", "paramforavdt", "rangemap", "cutoffmap",
     "exportmatlab", "exportRData", "verbose", "showerrors",
     "trysilent", "fracGTzero", "AIFshift", "slice",
     "IDvisit", "estpervoxeltlag")
     roiplotparams <- list(x_min, x_max, y_min, y_max,
     MAP_ul)
     names(roiplotparams) <- c("xmin", "xmax", "ymin",
     "ymax", "MAPul")
     dummy_data <- dcemri.data
     dcemri.data <- list(args, map_cc_slice, map_cc_roi,
     cc.median, map.times, aif, aif.shifted, mask.roi,
     map.KtransxT, map.KtransxT.cv, map.tlagxT,
     map.tlagxT.cv, map.kepxT, map.kepxT.cv, map.vbxT,
     map.vbxT.cv, map.vexT, map.OptimValuexT, map.fitfailuresxT,
     param.est.medianxT, roi.median.fitted.xTofts,
     param.est.whole.roi.xTofts, cv.whole.roi.xTofts,
     map.KtransT, map.KtransT.cv, map.kepT, map.kepT.cv,
     map.veT, map.OptimValueT, map.fitfailuresT,
     param.est.medianT, roi.median.fitted.Tofts,
     param.est.whole.roi.Tofts, proc.time.total,
     roiplotparams, KAT.version, map.AIC.xT, map.AIC.T,
     map.AIC.compare, nx, ny, nt, cc_fittedxT, cc_fittedT,
     p0.T, p0.xT, IRF.results, map.EF)
     names(dcemri.data) <- c("args", "cc", "ccroi",
     "ccmedian", "maptimes", "aif", "aifshifted",
     "maskroi", "mapKtransxT", "mapKtransxTcv",
     "maptlagxT", "maptlagxTcv", "mapkepxT", "mapkepxTcv",
     "mapvbxT", "mapvbxTcv", "mapvexT", "mapOptimValuexT",
     "mapfitfailuresxT", "paramestmedianxT", "roimedianfittedxTofts",
     "paramestwholeroixTofts", "cvwholeroixTofts",
     "mapKtransT", "mapKtransTcv", "mapkepT", "mapkepTcv",
     "mapveT", "mapOptimValueT", "mapfitfailuresT",
     "paramestmedianT", "roimedianfittedTofts",
     "paramestwholeroiTofts", "proctimetotal", "roiplotparams",
     "KATversion", "mapAICxT", "mapAICT", "mapAICcompare",
     "nx", "ny", "nt", "ccfittedxT", "ccfittedT",
     "p0T", "p0xT", "IRFresults", "mapEF")
     if (export.RData == TRUE) {
     cat("writing results to ", paste(results_file,
     ".RData", sep = ""), "...", sep = "", "\n")
     save(dcemri.data, file = paste(results_file,
     ".RData", sep = ""))
     }
     if (export.matlab == TRUE) {
     cat("writing results to ", paste(results_file,
     ".mat", sep = ""), "...", sep = "", "\n")
     writeMat(paste(results_file, ".mat", sep = ""),
     args = args, mapccslice = map_cc_slice, mapccroi = map_cc_roi,
     ccmedian = cc.median, maptimes = map.times,
     aif = aif, aifshifted = aif.shifted, maskroi = mask.roi,
     mapKtransxT = map.KtransxT, mapKtransxTcv = map.KtransxT.cv,
     maptlagxT = map.tlagxT, maptlagxTcv = map.tlagxT.cv,
     mapkepxT = map.kepxT, mapkepxTcv = map.kepxT.cv,
     mapvbxT = map.vbxT, mapvbxTcv = map.vbxT.cv,
     mapvexT = map.vexT, mapOptimValuexT = map.OptimValuexT,
     mapfitfailuresxT = map.fitfailuresxT, paramestmedianxT = param.est.medianxT,
     roimedianfittedxTofts = roi.median.fitted.xTofts,
     paramestwholeroixTofts = param.est.whole.roi.xTofts,
     cvwholeroixTofts = cv.whole.roi.xTofts, mapKtransT = map.KtransT,
     mapKtransTcv = map.KtransT.cv, mapkepT = map.kepT,
     mapkepTcv = map.kepT.cv, mapveT = map.veT,
     mapOptimValueT = map.OptimValueT, mapfitfailuresT = map.fitfailuresT,
     paramestmedianT = param.est.medianT, roimedianfittedTofts = roi.median.fitted.Tofts,
     paramestwholeroiTofts = param.est.whole.roi.Tofts,
     proctimetotal = proc.time.total, roiplotparams = roiplotparams,
     KATversion = KAT.version, mapAICxT = map.AIC.xT,
     mapAICT = map.AIC.T, mapAICcompare = map.AIC.compare,
     nx = nx, ny = ny, nt = nt, ccfittedxT = cc_fittedxT,
     ccfittedT = cc_fittedT, p0T = p0.T, p0xT = p0.xT,
     IRFresults = IRF.results, mapEF = map.EF)
     }
     dcemri.data <- dummy_data
     cat("..done in", format((proc.time()[3] - ptm)/60,
     digits = 2), "minutes.", "\n")
     if (export.matlab == FALSE)
     cat("Run KAT(filename.RData) to visualize results.",
     "\n")
     if (export.matlab == TRUE)
     cat("Run KAT(filename.RData) or KAT(filename.mat) to visualize results.",
     "\n")
     cat("--------", "\n")
     }
     }
     }
    }
    <bytecode: 0x48316a0>
    <environment: namespace:KATforDCEMRI>
     --- function search by body ---
    Function KAT in namespace KATforDCEMRI has this body.
     ----------- END OF FAILURE REPORT --------------
    Error in if (class(cov) != "try-error") { : the condition has length > 1
    Calls: demo ... withVisible -> eval -> eval -> runme -> system.time -> KAT
    Timing stopped at: 2.042 0.222 2.348
    Execution halted
Flavor: r-devel-linux-x86_64-fedora-clang

Version: 1.0.1
Check: examples
Result: ERROR
    Running examples in ‘KATforDCEMRI-Ex.R’ failed
    The error most likely occurred in:
    
    > ### Name: KAT
    > ### Title: Kinetic Analysis Tool for DCE-MRI
    > ### Aliases: KAT
    > ### Keywords: kinetic DCEMRI
    >
    > ### ** Examples
    >
    > ## Create temporary directory for example code output files
    > temp_dir <- tempdir(check=FALSE)
    > ##
    > current_dir <- getwd()
    > setwd(temp_dir)
    > ##
    > ## Run example code
    > demo(KAT, ask=FALSE)
    
    
     demo(KAT)
     ---- ~~~
    
    > ## KATforDCEMRI: a Kinetic Analysis Tool for DCE-MRI
    > ## Copyright 2018 Genentech, Inc.
    > ##
    > ## For questions or comments, please contact
    > ## Gregory Z. Ferl, Ph.D.
    > ## Genentech, Inc.
    > ## Development Sciences
    > ## 1 DNA Way, Mail stop 463A
    > ## South San Francisco, CA, United States of America
    > ## E-mail: ferl.gregory@gene.com
    >
    > runme <- function(){
    + data(dcemri.data, package="KATforDCEMRI")
    +
    + ## dir.create("KATforDCEMRI_benchmark_test")
    + ## setwd("KATforDCEMRI_benchmark_test")
    +
    + attach(dcemri.data)
    +
    + ## SHRINK THE ROI MASK
    + maskROI[,,] <- 0
    + #maskROI[32:42,32:42,] <- 1
    + maskROI[34:36,34:36,] <- 1
    +
    + runtime1 <- system.time(KAT.checkData(file.name="KAT", vector.times=vectorTimes, map.CC=mapCC, mask.ROI=maskROI, vector.AIF=vectorAIF))
    + runtime2 <- system.time(KAT(file = "KAT.RData", results_file="KAT_benchmark_test-full", range.map=1.05, cutoff.map=0.95, AIF.shift="NONE", tlag.Tofts.on=FALSE, export.matlab=FALSE))
    +
    + ## runtime3 <- system.time(KAT.checkData(file.name="KATtrunc", vector.times=vectorTimes[1:44], map.CC=mapCC[,,,1:44], mask.ROI=maskROI, vector.AIF=vectorAIF[1:44]))
    + ## runtime4 <- system.time(KAT(file = "KATtrunc.RData", results_file="KAT_benchmark_test-truncated", range.map=1.05, cutoff.map=0.95))
    + detach(dcemri.data)
    +
    + ## runtime <- format(runtime1[3] + runtime2[3] + runtime3[3] + runtime4[3], digits=3)
    + runtime <- format(runtime1[3] + runtime2[3], digits=3)
    +
    + ## KAT.plot(F1="KAT_benchmark_test-full_slice1.RData", F2="KAT_benchmark_test-full_slice2.RData", F3="KAT_benchmark_test-truncated_slice1.RData", F4="KAT_benchmark_test-truncated_slice2.RData", export.matlab=FALSE)
    + KAT.plot(F1="KAT_benchmark_test-full_slice1.RData", F2="KAT_benchmark_test-full_slice2.RData", F3="KAT_benchmark_test-full_slice3.RData", F4="KAT_benchmark_test-full_slice4.RData", export.matlab=FALSE)
    +
    + load("KAT_benchmark_test-full_slice1.RData")
    + Ktrans_s1A <- as.numeric(format(dcemri.data$paramestmedianxT$Ktrans.median, digits=3))
    + kep_s1A <- as.numeric(format(dcemri.data$paramestmedianxT$kep.median, digits=3))
    + vb_s1A <- as.numeric(format(dcemri.data$paramestmedianxT$vb.median, digits=3))
    + cvKtrans_s1A <- as.numeric(format(100*sd(as.vector(dcemri.data$mapKtransxT), na.rm=TRUE)/Ktrans_s1A, digits=3))
    + cvkep_s1A <- as.numeric(format(100*sd(as.vector(dcemri.data$mapkepxT), na.rm=TRUE)/kep_s1A, digits=3))
    + cvvb_s1A <- as.numeric(format(100*sd(as.vector(dcemri.data$mapvbxT), na.rm=TRUE)/vb_s1A, digits=3))
    +
    + Ktrans_Ts1A <- as.numeric(format(dcemri.data$paramestmedianT$Ktrans.median, digits=3))
    + kep_Ts1A <- as.numeric(format(dcemri.data$paramestmedianT$kep.median, digits=3))
    + cvKtrans_Ts1A <- as.numeric(format(100*sd(as.vector(dcemri.data$mapKtransT), na.rm=TRUE)/Ktrans_Ts1A, digits=3))
    + cvkep_Ts1A <- as.numeric(format(100*sd(as.vector(dcemri.data$mapkepT), na.rm=TRUE)/kep_s1A, digits=3))
    +
    + load("KAT_benchmark_test-full_slice2.RData")
    + Ktrans_s2A <- as.numeric(format(dcemri.data$paramestmedianxT$Ktrans.median, digits=3))
    + kep_s2A <- as.numeric(format(dcemri.data$paramestmedianxT$kep.median, digits=3))
    + vb_s2A <- as.numeric(format(dcemri.data$paramestmedianxT$vb.median, digits=3))
    + cvKtrans_s2A <- as.numeric(format(100*sd(as.vector(dcemri.data$mapKtransxT), na.rm=TRUE)/Ktrans_s2A, digits=3))
    + cvkep_s2A <- as.numeric(format(100*sd(as.vector(dcemri.data$mapkepxT), na.rm=TRUE)/kep_s2A, digits=3))
    + cvvb_s2A <- as.numeric(format(100*sd(as.vector(dcemri.data$mapvbxT), na.rm=TRUE)/vb_s2A, digits=3))
    +
    + Ktrans_Ts2A <- as.numeric(format(dcemri.data$paramestmedianT$Ktrans.median, digits=3))
    + kep_Ts2A <- as.numeric(format(dcemri.data$paramestmedianT$kep.median, digits=3))
    + cvKtrans_Ts2A <- as.numeric(format(100*sd(as.vector(dcemri.data$mapKtransT), na.rm=TRUE)/Ktrans_Ts2A, digits=3))
    + cvkep_Ts2A <- as.numeric(format(100*sd(as.vector(dcemri.data$mapkepT), na.rm=TRUE)/kep_s2A, digits=3))
    +
    + ## load("KAT_benchmark_test-truncated_slice1.RData")
    + load("KAT_benchmark_test-full_slice3.RData")
    + Ktrans_s1B <- as.numeric(format(dcemri.data$paramestmedianxT$Ktrans.median, digits=3))
    + kep_s1B <- as.numeric(format(dcemri.data$paramestmedianxT$kep.median, digits=3))
    + vb_s1B <- as.numeric(format(dcemri.data$paramestmedianxT$vb.median, digits=3))
    + cvKtrans_s1B <- as.numeric(format(100*sd(as.vector(dcemri.data$mapKtransxT), na.rm=TRUE)/Ktrans_s1B, digits=3))
    + cvkep_s1B <- as.numeric(format(100*sd(as.vector(dcemri.data$mapkepxT), na.rm=TRUE)/kep_s1B, digits=3))
    + cvvb_s1B <- as.numeric(format(100*sd(as.vector(dcemri.data$mapvbxT), na.rm=TRUE)/vb_s1B, digits=3))
    +
    + Ktrans_Ts1B <- as.numeric(format(dcemri.data$paramestmedianT$Ktrans.median, digits=3))
    + kep_Ts1B <- as.numeric(format(dcemri.data$paramestmedianT$kep.median, digits=3))
    + cvKtrans_Ts1B <- as.numeric(format(100*sd(as.vector(dcemri.data$mapKtransT), na.rm=TRUE)/Ktrans_Ts1B, digits=3))
    + cvkep_Ts1B <- as.numeric(format(100*sd(as.vector(dcemri.data$mapkepT), na.rm=TRUE)/kep_s1B, digits=3))
    +
    + ## load("KAT_benchmark_test-truncated_slice2.RData")
    + load("KAT_benchmark_test-full_slice4.RData")
    + Ktrans_s2B <- as.numeric(format(dcemri.data$paramestmedianxT$Ktrans.median, digits=3))
    + kep_s2B <- as.numeric(format(dcemri.data$paramestmedianxT$kep.median, digits=3))
    + vb_s2B <- as.numeric(format(dcemri.data$paramestmedianxT$vb.median, digits=3))
    + cvKtrans_s2B <- as.numeric(format(100*sd(as.vector(dcemri.data$mapKtransxT), na.rm=TRUE)/Ktrans_s2B, digits=3))
    + cvkep_s2B <- as.numeric(format(100*sd(as.vector(dcemri.data$mapkepxT), na.rm=TRUE)/kep_s2B, digits=3))
    + cvvb_s2B <- as.numeric(format(100*sd(as.vector(dcemri.data$mapvbxT), na.rm=TRUE)/vb_s2B, digits=3))
    +
    + Ktrans_Ts2B <- as.numeric(format(dcemri.data$paramestmedianT$Ktrans.median, digits=3))
    + kep_Ts2B <- as.numeric(format(dcemri.data$paramestmedianT$kep.median, digits=3))
    + cvKtrans_Ts2B <- as.numeric(format(100*sd(as.vector(dcemri.data$mapKtransT), na.rm=TRUE)/Ktrans_Ts2B, digits=3))
    + cvkep_Ts2B <- as.numeric(format(100*sd(as.vector(dcemri.data$mapkepT), na.rm=TRUE)/kep_s2B, digits=3))
    +
    + pdf(file="KAT_demo-page1.pdf", height=11, width=8.5)
    +
    + plot(0:35, type = "n", xaxt="n", yaxt="n", bty="n", xlab = "", ylab = "")
    + text(5, 35, paste("KATforDCEMRI version ", dcemri.data$KATversion, " BENCHMARK TEST", sep=""), pos=4, font=2, col="red")
    + text(5, 33, paste("date:", date(), "\n"), pos=4)
    + text(5, 32, paste("Total Processing Time:", runtime, "seconds"), pos=4)
    + text(5, 29, paste("sysname/release:", Sys.info()[[1]], Sys.info()[[2]], "\n"), pos=4)
    + text(5, 28, paste("nodename:", Sys.info()[[4]], "\n"), pos=4)
    + text(5, 27, paste("user:", Sys.info()[[7]], "\n"), pos=4)
    + text(5, 25, "Estimated (inter-voxel %CV) versus True Parameter values (extended Tofts)", pos=4, font=2)
    + text(5, 23, "LOW NOISE DATA", pos=4)
    + text(5, 22, paste("Slice 1: Ktrans =", Ktrans_s1A, "1/min (", cvKtrans_s1A,"%) [true value = 0.22 1/min]", sep=""), pos=4)
    + text(5, 21, paste("Slice 1: kep =", kep_s1A, "1/min (", cvkep_s1A, "%) [true value = 1.1 1/min]", sep=""), pos=4)
    + text(5, 20, paste("Slice 1: vb =", vb_s1A, " (", cvvb_s1A, "%) [true value = 0]", sep=""), pos=4)
    + text(5, 18, paste("Slice 2: Ktrans =", Ktrans_s2A, "1/min (", cvKtrans_s2A,"%) [true value = 0.22 1/min]", sep=""), pos=4)
    + text(5, 17, paste("Slice 2: kep =", kep_s2A, "1/min (", cvkep_s2A, "%) [true value = 1.1 1/min]", sep=""), pos=4)
    + text(5, 16, paste("Slice 2: vb =", vb_s2A, " (", cvvb_s2A, "%) [true value = 0.05]", sep=""), pos=4)
    + text(5, 14, "HIGH NOISE DATA", pos=4)
    + text(5, 13, paste("Slice 3: Ktrans =", Ktrans_s1B, "1/min (", cvKtrans_s1B,"%) [true value = 0.22 1/min]", sep=""), pos=4)
    + text(5, 12, paste("Slice 3: kep =", kep_s1B, "1/min (", cvkep_s1B, "%) [true value = 1.1 1/min]", sep=""), pos=4)
    + text(5, 11, paste("Slice 3: vb =", vb_s1B, " (", cvvb_s1B, "%) [true value = 0]", sep=""), pos=4)
    + text(5, 9, paste("Slice 4: Ktrans =", Ktrans_s2B, "1/min (", cvKtrans_s2B,"%) [true value = 0.22 1/min]", sep=""), pos=4)
    + text(5, 8, paste("Slice 4: kep =", kep_s2B, "1/min (", cvkep_s2B, "%) [true value = 1.1 1/min]", sep=""), pos=4)
    + text(5, 7, paste("Slice 4: vb =", vb_s2B, " (", cvvb_s2B, "%) [true value = 0.05]", sep=""), pos=4)
    + dev.off()
    +
    +
    + pdf(file="KAT_demo-page2.pdf", height=11, width=8.5)
    +
    + plot(0:35, type = "n", xaxt="n", yaxt="n", bty="n", xlab = "", ylab = "")
    +
    + text(5, 25, "Estimated (inter-voxel %CV) versus True Parameter values (Tofts Model)", pos=4, font=2)
    + text(5, 23, "LOW NOISE DATA", pos=4)
    + text(5, 22, paste("Slice 1: Ktrans =", Ktrans_Ts1A, "1/min (", cvKtrans_Ts1A,"%) [true value = 0.22 1/min]", sep=""), pos=4)
    + text(5, 21, paste("Slice 1: kep =", kep_Ts1A, "1/min (", cvkep_Ts1A, "%) [true value = 1.1 1/min]", sep=""), pos=4)
    + text(5, 20, "Slice 1: vb = 0 (Fixed) [true value = 0]", pos=4)
    + text(5, 18, paste("Slice 2: Ktrans =", Ktrans_Ts2A, "1/min (", cvKtrans_Ts2A,"%) [true value = 0.22 1/min]", sep=""), pos=4)
    + text(5, 17, paste("Slice 2: kep =", kep_Ts2A, "1/min (", cvkep_Ts2A, "%) [true value = 1.1 1/min]", sep=""), pos=4)
    + text(5, 16, "Slice 2: vb = 0 (Fixed) [true value = 0.05]", pos=4)
    + text(5, 14, "HIGH NOISE DATA", pos=4)
    + text(5, 13, paste("Slice 3: Ktrans =", Ktrans_Ts1B, "1/min (", cvKtrans_Ts1B,"%) [true value = 0.22 1/min]", sep=""), pos=4)
    + text(5, 12, paste("Slice 3: kep =", kep_Ts1B, "1/min (", cvkep_Ts1B, "%) [true value = 1.1 1/min]", sep=""), pos=4)
    + text(5, 11, "Slice 3: vb = 0 (Fixed) [true value = 0]", pos=4)
    + text(5, 9, paste("Slice 4: Ktrans =", Ktrans_Ts2B, "1/min (", cvKtrans_Ts2B,"%) [true value = 0.22 1/min]", sep=""), pos=4)
    + text(5, 8, paste("Slice 4: kep =", kep_Ts2B, "1/min (", cvkep_Ts2B, "%) [true value = 1.1 1/min]", sep=""), pos=4)
    + text(5, 7, "Slice 4: vb = 0 (Fixed) [true value = 0.05]", pos=4)
    + dev.off()
    + }
    
    > runme()
    
    checking dimensions of vectors and arrays...
    
    length of vector.times is 89 with units of seconds
    length of vector.AIF is 89
    dimensions of map.CC array are 75 x 75 x 4 slices x 89 time points
    dimensions of mask.ROI array are 75 x 75 x 4 slices
    
    ...vector and array dimensions are okay.
    
    Saving data in a single R file...
    ...file saved as KAT.RData ...
    ...use the KAT() function to analyze data within this file.
    
    
    #########################################################################
    ##-------------- KINETIC ANALYSIS TOOL (KAT) FOR DCE-MRI --------------##
    #########################################################################
    ##---------------------- R package version 1.0 ----------------------##
    #########################################################################
    
    loading KAT.RData into R...
    ..done in 0.0017 minutes.
    --------
    ***** ROI DETECTED IN SLICE 1 *****
    --------
    extracting slice 1 for analysis...
    ..done in 0.004 minutes.
    --------
    applying ROI mask to cc matrix...
    ..done in 7e-04 minutes.
    --------
    fitting xTofts and Tofts models to whole ROI data...
    ..done in 0.004 minutes.
    --------
     ----------- FAILURE REPORT --------------
     --- failure: the condition has length > 1 ---
     --- srcref ---
    :
     --- package (from environment) ---
    KATforDCEMRI
     --- call from context ---
    KAT(file = "KAT.RData", results_file = "KAT_benchmark_test-full",
     range.map = 1.05, cutoff.map = 0.95, AIF.shift = "NONE",
     tlag.Tofts.on = FALSE, export.matlab = FALSE)
     --- call from argument ---
    if (class(cov) != "try-error") {
     sd <- try(sqrt(diag(cov)), silent = try.silent)
     if (class(sd) != "try-error") {
     cv <- sd/param_est * 100
     cv.whole.roi.xTofts <- as.numeric(format(cv, digits = 1))
     }
    }
     --- R stacktrace ---
    where 1 at /data/gannet/ripley/R/packages/tests-devel/KATforDCEMRI.Rcheck/KATforDCEMRI/demo/KAT.R#26: KAT(file = "KAT.RData", results_file = "KAT_benchmark_test-full",
     range.map = 1.05, cutoff.map = 0.95, AIF.shift = "NONE",
     tlag.Tofts.on = FALSE, export.matlab = FALSE)
    where 2 at /data/gannet/ripley/R/packages/tests-devel/KATforDCEMRI.Rcheck/KATforDCEMRI/demo/KAT.R#26: system.time(KAT(file = "KAT.RData", results_file = "KAT_benchmark_test-full",
     range.map = 1.05, cutoff.map = 0.95, AIF.shift = "NONE",
     tlag.Tofts.on = FALSE, export.matlab = FALSE))
    where 3 at /data/gannet/ripley/R/packages/tests-devel/KATforDCEMRI.Rcheck/KATforDCEMRI/demo/KAT.R#141: runme()
    where 4: eval(ei, envir)
    where 5: eval(ei, envir)
    where 6: withVisible(eval(ei, envir))
    where 7: source(available, echo = echo, max.deparse.length = Inf, keep.source = TRUE,
     encoding = encoding)
    where 8: demo(KAT, ask = FALSE)
    
     --- value of length: 2 type: logical ---
    [1] TRUE TRUE
     --- function from context ---
    function (file = "concatenate.KAT.with.KAT.checkData.RData",
     results_file = "my_results", method.optimization = "L-BFGS-B",
     show.rt.fits = FALSE, param.for.avdt = "Ktrans", range.map = 1.5,
     cutoff.map = 0.85, export.matlab = TRUE, export.RData = TRUE,
     verbose = FALSE, show.errors = FALSE, try.silent = TRUE,
     fracGTzero = 0.75, AIF.shift = "", Force.AIF.peak = FALSE,
     tlag.Tofts.on = FALSE, est.per.voxel.tlag = FALSE, ...)
    {
     lo <- 0
     options(show.error.messsages = show.errors)
     ftype <- strsplit(file, split = "a")[[1]]
     ftype <- ftype[length(ftype) - 1]
     ftype <- strsplit(ftype, split = "")[[1]]
     ftype <- ftype[length(ftype)]
     if (ftype == "m")
     file.format <- "matlab"
     if (ftype == "D")
     file.format <- "RData"
     file_short <- file
     KAT.version <- "1.0"
     ptm_total <- proc.time()[3]
     modeltype1 <- "xTofts"
     modeltype2 <- "Tofts"
     funcz <- function(x) {
     as.numeric(as.character(x))
     }
     aif.shift.func <- function(t, cp, time_shift) {
     x <- cbind(t, cp)
     tmax <- subset(x[, 1], x[, 2] == max(x[, 2]))
     if (AIF.shift == "ARTERY")
     tmax <- tmax + time_shift
     if (AIF.shift == "VEIN")
     tmax <- tmax - time_shift
     cpFUNC <- approxfun(t, cp, rule = 2)
     if (AIF.shift == "ARTERY")
     tshift <- t - time_shift
     if (AIF.shift == "VEIN")
     tshift <- t + time_shift
     cp.shift <- cpFUNC(tshift)
     if (Force.AIF.peak == TRUE)
     cp.shift[cp.shift == max(cp.shift)] <- max(cp)
     return(cp.shift)
     }
     roi.modelT <- function(p, t, dt, cp, zing = 0) {
     if (zing == 0) {
     Ktrans <- p[1]
     kep <- p[2]
     if (tlag.Tofts.on == TRUE) {
     if (fix.tlag != TRUE & AIF.shift != "NONE")
     time_shift <- p[3]
     if (fix.tlag == TRUE & AIF.shift != "NONE")
     time_shift <- param.est.whole.roi.Tofts$tlag
     }
     if (tlag.Tofts.on == FALSE) {
     time_shift <- param.est.whole.roi.xTofts$tlag
     }
     if (AIF.shift == "ARTERY" | AIF.shift == "VEIN")
     cp <- aif.shift.func(t, cp, time_shift)
     f <- Ktrans * exp(kep * t) * cp
     int <- c(0, cumsum(dt * (tail(f, -1) + head(f, -1)))/2)
     ct <- exp(-kep * t) * int
     return(ct)
     }
     if (zing == 1)
     return("modelT")
     }
     roi.modelxT <- function(p, t, dt, cp, zing = 0) {
     if (zing == 0) {
     Ktrans <- p[1]
     kep <- p[2]
     vb <- p[3]
     if (fix.tlag != TRUE & AIF.shift != "NONE")
     time_shift <- p[4]
     if (est.per.voxel.tlag == FALSE) {
     if (fix.tlag == TRUE & AIF.shift != "NONE")
     time_shift <- param.est.whole.roi.xTofts$tlag
     }
     if (est.per.voxel.tlag == TRUE) {
     if (fix.tlag == TRUE & AIF.shift != "NONE")
     time_shift <- p[4]
     }
     if (AIF.shift == "ARTERY" | AIF.shift == "VEIN")
     cp <- aif.shift.func(t, cp, time_shift)
     f <- Ktrans * exp(kep * t) * cp
     int <- c(0, cumsum(dt * (tail(f, -1) + head(f, -1)))/2)
     ct <- exp(-kep * t) * int
     ct <- ct + vb * cp
     return(ct)
     }
     if (zing == 1)
     return("modelxT")
     }
     calch <- function(u, y, TIME_trunc) {
     locfit_y <- preplot(y, newdata = 0:max(TIME_trunc))
     y_smooth <- locfit_y$fit
     u <- u[match(u[u == max(u)], u):length(u)]
     y_smooth <- y_smooth[match(u[u == max(u)], u):length(u)]
     n <- length(u)
     A <- matrix(0, nrow = n, ncol = n)
     ind <- row(A) - col(A)
     ind[ind < 0] <- (-1)
     ind <- ind + 2
     A <- matrix(c(0, u)[ind], nrow = n, ncol = n)
     h <- solve(A, y_smooth)
     h.time.vector <- (1:length(h)) - 1
     out <- list(h.time.vector, h)
     names(out) <- c("t", "IRF")
     return(out)
     }
     calchFUNC <- function(vector.times, AIF, map_cc_slice, correct.vp = TRUE,
     alpha.AIF = c(0.1, 0.5), vp.nom = 0.1, kep.nom = 0.5) {
     AUMC <- function(AUMC.median, h.median, irf_time_vec,
     r) {
     AUMC.median <- AUMC.median + 0.5 * (h.median[r] *
     irf_time_vec[r] + h.median[r + 1] * irf_time_vec[r +
     1])
     }
     artery_data <- data.frame(vector.times * 60, AIF)
     names(artery_data) <- c("TIME", "ARTERY")
     data_artery_peak <- subset(artery_data, artery_data$ARTERY ==
     max(artery_data$ARTERY))
     data_remove_artery_prepeak <- subset(artery_data, artery_data$TIME >=
     data_artery_peak$TIME)
     frames_to_peak <- length(artery_data[, 1]) - length(data_remove_artery_prepeak[,
     1]) + 1
     TIME <- data_remove_artery_prepeak$TIME
     ARTERY <- data_remove_artery_prepeak$ARTERY
     TIME_trunc <- TIME[seq(1, length(TIME) - 1, by = 1)]
     TIME_trunc <- TIME_trunc - TIME_trunc[1]
     ARTERY_trunc <- ARTERY[seq(1, length(ARTERY) - 1, by = 1)]
     ARTERY_smooth <- locfit.robust(ARTERY_trunc ~ TIME_trunc,
     acri = "cp", alpha = alpha.AIF)
     AIF_smooth <- ARTERY_smooth
     locfit_u <- preplot(AIF_smooth, newdata = 0:max(TIME_trunc))
     u_smooth <- locfit_u$fit
     Tmax <- max(TIME_trunc)
     TUMOR.median <- map_cc_slice
     TUMOR.median <- TUMOR.median[seq(frames_to_peak, length(TUMOR.median),
     by = 1)]
     if (vp.nom > 0)
     TUMOR.median_corr <- TUMOR.median - vp.nom * ARTERY
     if (vp.nom <= 0)
     TUMOR.median_corr <- TUMOR.median
     TUMOR.median_corr_shifted <- TUMOR.median_corr[seq(2,
     length(TUMOR.median_corr), by = 1)]
     TUMOR.median_smooth <- locfit.robust(TUMOR.median_corr_shifted ~
     TIME_trunc, acri = "cp")
     calch.out <- calch(u_smooth, TUMOR.median_smooth, TIME_trunc)
     h.median <- calch.out$IRF
     irf_time_vec <- calch.out$t
     n <- length(h.median)
     AUC.median <- 0
     AUMC.median <- 0
     for (r in 1:(n - 1)) {
     h_sum <- h.median[r] + h.median[r + 1]
     t_sum <- irf_time_vec[r] + irf_time_vec[r + 1]
     AUC.median <- AUC.median + 0.5 * h_sum
     AUMC.median <- AUMC(AUMC.median, h.median, irf_time_vec,
     r)
     }
     AUCMRT.median <- AUC.median/(AUMC.median/AUC.median) *
     60
     if (kep.nom > 0) {
     t_scan <- max(TIME_trunc)/60
     ve_trunc_error <- 1 - exp(-kep.nom * t_scan)
     Ktrans_trunc_error <- (1 - exp(-kep.nom * t_scan))^2/(1 -
     (1 + kep.nom * t_scan) * exp(-kep.nom * t_scan))
     AUC.median <- AUC.median/ve_trunc_error
     AUCMRT.median <- AUCMRT.median/Ktrans_trunc_error
     }
     irf_time_vec <- irf_time_vec/60
     out <- list(AUC.median, AUCMRT.median, h.median, irf_time_vec)
     names(out) <- c("AUCh", "AUChMRTh", "IRF", "t")
     return(out)
     }
     Obj_roi <- function(p, model, t, dt, cp, roi) {
     sum((model(p, t, dt, cp) - roi)^2)
     }
     map_cc_slice <- NULL
     map_cc_roi <- NULL
     aif <- NULL
     aif.shifted <- NULL
     map.times <- NULL
     map.KtransxT <- NULL
     map.kepxT <- NULL
     map.vexT <- NULL
     map.vbxT <- NULL
     map.tlagxT <- NULL
     map.fitfailuresxT <- NULL
     map.KtransT <- NULL
     map.kepT <- NULL
     map.veT <- NULL
     map.fitfailuresT <- NULL
     mask.roi <- NULL
     nx <- NULL
     ny <- NULL
     nt <- NULL
     map.KtransxT <- NULL
     map.kepxT <- NULL
     map.vexT <- NULL
     map.vbxT <- NULL
     map.AIC.compare <- NULL
     map.AIC.T <- NULL
     map.EF <- NULL
     map.AIC.xT <- NULL
     roi.median.fitted.Tofts <- NULL
     param.est.whole.roi.Tofts <- NULL
     roi.median.fitted.xTofts <- NULL
     param.est.whole.roi.xTofts <- NULL
     cv.whole.roi.xTofts <- NULL
     cat("\n")
     cat("#########################################################################",
     "\n")
     cat("##-------------- KINETIC ANALYSIS TOOL (KAT) FOR DCE-MRI --------------##",
     "\n")
     cat("#########################################################################",
     "\n")
     cat("##---------------------- R package version", KAT.version,
     "----------------------##", "\n")
     cat("#########################################################################",
     "\n")
     cat("\n")
     filea <- strsplit(file, split = "/")[[1]]
     fileb <- filea[length(filea)]
     if (file != "concatenate.KAT.with.KAT.checkData.RData")
     cat("loading", fileb, "into R...", "\n")
     ptm <- proc.time()[3]
     if (file.format == "matlab") {
     mat_data <- readMat(file)
     if (length(names(mat_data)) < 10)
     file.original <- TRUE
     if (length(names(mat_data)) >= 10)
     file.original <- FALSE
     }
     if (file.format == "RData") {
     if (file != "concatenate.KAT.with.KAT.checkData.RData")
     load(file)
     if (length(names(dcemri.data)) < 10) {
     file.original <- TRUE
     ROIcounter <- apply(dcemri.data$maskROI, 3, max)
     results_file_temp <- results_file
     }
     if (length(names(dcemri.data)) >= 10) {
     file.original <- FALSE
     ROIcounter <- 1
     }
     }
     if (file != "concatenate.KAT.with.KAT.checkData.RData") {
     cat("..done in", format((proc.time()[3] - ptm)/60, digits = 2),
     "minutes.", "\n")
     cat("--------", "\n")
     }
     for (slicenumber in 1:(length(ROIcounter))) {
     if (ROIcounter[slicenumber] == 1) {
     if (file.original == TRUE) {
     slice <- slicenumber
     cat("***** ROI DETECTED IN SLICE", slice, " *****",
     "\n")
     cat("--------", "\n")
     }
     if (file.original == TRUE) {
     if (AIF.shift != "VEIN" & AIF.shift != "ARTERY" &
     AIF.shift != "NONE")
     stop("You must specify the argument AIF.shift argument as VEIN, ARTERY or NONE, indicating that the AIF you are using is based on data from a vein or artery or NONE if tlag should be set to 0. This will ensure that the time lag parameter in the Tofts and xTofts models has the appropriate inital value and is bounded correctly; either -Inf to 0 (for VEIN) or 0 to Inf (for ARTERY).")
     filenameTag <- paste("_slice", slice, sep = "")
     results_file <- paste(results_file_temp, filenameTag,
     sep = "")
     roi.model <- roi.modelxT
     if (slice == "" || slice < 0)
     stop("The slice argument has not been properly specified; slice=``slice number''")
     cat("extracting slice", slice, "for analysis...",
     "\n")
     ptm <- proc.time()[3]
     if (file.format == "matlab") {
     map.times <- as.vector(mat_data$map[[4]]/60)
     map_cc <- mat_data$map[[3]]
     map_cc_slice <- map_cc[, , slice, ]
     if (length(map.times) != length(map_cc_slice[1,
     1, ]))
     stop("The length of the time vector does not match the length of the contrast agent concentration vector")
     }
     if (file.format == "RData") {
     map.times <- as.vector(dcemri.data$vectorTimes/60)
     map_cc <- dcemri.data$mapCC
     map_cc_slice <- map_cc[, , slice, ]
     if (length(map.times) != length(map_cc_slice[1,
     1, ]))
     stop("The length of the time vector does not match the length of the contrast agent concentration vector")
     }
     nt <- length(map_cc_slice[1, 1, ])
     ny <- length(map_cc_slice[, 1, 1])
     nx <- length(map_cc_slice[1, , 1])
     ccTEMP <- rep(0, prod(dim(map_cc_slice)))
     dim(ccTEMP) <- dim(map_cc_slice)[c(2, 1, 3)]
     for (i in 1:nt) ccTEMP[, , i] <- rot90(map_cc_slice[,
     , i], 3)
     map_cc_slice <- ccTEMP
     if (file.format == "matlab")
     mask.roi <- mat_data$mask[[1]][, , slice]
     if (file.format == "RData")
     mask.roi <- dcemri.data$mask[, , slice]
     mask.roi <- rot90(mask.roi, 3)
     if (max(mask.roi) != 1)
     stop("Your ROI mask is either composed entirely of zeroes or contains nonnumeric elements; voxels within the ROI should have a value of ``1'' and all other voxels should have a value of ``0''.")
     if (file.format == "matlab")
     aif <- as.vector(mat_data$aif)
     if (file.format == "RData")
     aif <- as.vector(dcemri.data$vectorAIF)
     if (file.format == "matlab")
     rm(mat_data)
     cat("..done in", format((proc.time()[3] - ptm)/60,
     digits = 2), "minutes.", "\n")
     cat("--------", "\n")
     cat("applying ROI mask to cc matrix...", "\n")
     ptm <- proc.time()[3]
     map_cc_roi <- map_cc_slice
     mask.roi.temp <- mask.roi
     mask.roi.temp[mask.roi.temp == 0] <- NA
     for (z in 1:nt) map_cc_roi[, , z] <- map_cc_roi[,
     , z] * mask.roi.temp
     cat("..done in", format((proc.time()[3] - ptm)/60,
     digits = 2), "minutes.", "\n")
     cat("--------", "\n")
     ptm <- proc.time()[3]
     cc.median <- seq(1:nt)
     for (t in 1:nt) cc.median[t] <- median(map_cc_roi[,
     , t], na.rm = TRUE)
     cat("fitting", modeltype1, "and", modeltype2,
     "models to whole ROI data...", "\n")
     IRF.out <- calchFUNC(map.times, aif, cc.median)
     AUCcorrnom <- IRF.out$AUCh
     AUCMRTcorrnom <- IRF.out$AUChMRTh
     AUCMRTcorrnom.divby.AUCcorrnom <- IRF.out$AUChMRTh/IRF.out$AUCh
     if (tlag.Tofts.on == TRUE & AIF.shift == "VEIN") {
     p0.T.median <- c(IRF.out$AUChMRTh, IRF.out$AUChMRTh/IRF.out$AUCh,
     0.05)
     lower.wholeT <- c(lo, lo, 0)
     upper.wholeT <- c(Inf, Inf, Inf)
     }
     if (tlag.Tofts.on == TRUE & AIF.shift == "ARTERY") {
     p0.T.median <- c(IRF.out$AUChMRTh, IRF.out$AUChMRTh/IRF.out$AUCh,
     0.05)
     lower.wholeT <- c(lo, lo, 0)
     upper.wholeT <- c(Inf, Inf, Inf)
     }
     if (tlag.Tofts.on == FALSE) {
     p0.T.median <- c(IRF.out$AUChMRTh, IRF.out$AUChMRTh/IRF.out$AUCh)
     lower.wholeT <- c(lo, lo)
     upper.wholeT <- c(Inf, Inf)
     }
     if (AIF.shift == "VEIN") {
     p0.xT.median <- c(IRF.out$AUChMRTh, IRF.out$AUChMRTh/IRF.out$AUCh,
     0.05, 0.05)
     lower.wholexT <- c(lo, lo, lo, 0)
     upper.wholexT <- c(Inf, Inf, Inf, Inf)
     }
     if (AIF.shift == "ARTERY") {
     p0.xT.median <- c(IRF.out$AUChMRTh, IRF.out$AUChMRTh/IRF.out$AUCh,
     0.05, 0.05)
     lower.wholexT <- c(lo, lo, lo, 0)
     upper.wholexT <- c(Inf, Inf, Inf, Inf)
     }
     if (AIF.shift == "NONE") {
     p0.xT.median <- c(IRF.out$AUChMRTh, IRF.out$AUChMRTh/IRF.out$AUCh,
     0.05)
     lower.wholexT <- c(lo, lo, lo)
     upper.wholexT <- c(Inf, Inf, Inf)
     }
     SAAMII <- FALSE
     if (SAAMII == TRUE) {
     hw.x <- cbind(format(map.times, digits = 3),
     format(aif, digits = 3), format(cc.median,
     digits = 3))
     write.table("DATA", file = paste("slice", slice,
     "_forSAAMII", sep = ""), quote = FALSE, row.names = FALSE,
     col.names = FALSE)
     write.table("(SD 1)", file = paste("slice",
     slice, "_forSAAMII", sep = ""), quote = FALSE,
     row.names = FALSE, append = TRUE, col.names = FALSE)
     write.table(hw.x, file = paste("slice", slice,
     "_forSAAMII", sep = ""), quote = FALSE, row.names = FALSE,
     append = TRUE, col.names = c("t", paste("AIF_s",
     slice, sep = ""), paste("CC_s", slice,
     sep = "")))
     write.table("END", file = paste("slice", slice,
     "_forSAAMII", sep = ""), quote = FALSE, row.names = FALSE,
     append = TRUE, col.names = FALSE)
     }
     roi <- cc.median
     t <- map.times
     fix.tlag <- FALSE
     roi.model <- roi.modelxT
     if (method.optimization == "L-BFGS-B")
     fit.roi.median.xTofts <- try(optim(p0.xT.median,
     Obj_roi, model = roi.model, t = map.times,
     dt = diff(map.times), cp = aif, roi = roi,
     method = "L-BFGS-B", lower = lower.wholexT,
     upper = upper.wholexT, hessian = TRUE), silent = try.silent)
     if (method.optimization != "L-BFGS-B")
     fit.roi.median.xTofts <- try(optim(p0.xT.median,
     Obj_roi, model = roi.model, t = map.times,
     dt = diff(map.times), cp = aif, roi = roi,
     method = method.optimization, hessian = TRUE),
     silent = try.silent)
     if (class(fit.roi.median.xTofts) != "try-error") {
     if (fit.roi.median.xTofts$convergence == 0) {
     roi.median.fitted.xTofts <- roi.model(p = fit.roi.median.xTofts$par,
     t = map.times, dt = diff(map.times), cp = aif)
     params.xTofts <- fit.roi.median.xTofts$par
     param.est.whole.roi.xTofts <- list(fit.roi.median.xTofts$par[1],
     fit.roi.median.xTofts$par[2], fit.roi.median.xTofts$par[3],
     fit.roi.median.xTofts$par[4])
     names(param.est.whole.roi.xTofts) <- c("Ktrans",
     "kep", "vb", "tlag")
     }
     }
     if (class(fit.roi.median.xTofts) == "try-error") {
     cat("A problem occured when fitting the extended Tofts model to median intensity/concentration data across the ROI. Optimization algorithm did not converge. This may occur when a large fraction of voxels within the ROI are non-enhancing, as these are not excluded from analysis of the median intensity/concentration profile",
     "\n")
     }
     if (class(fit.roi.median.xTofts) != "try-error") {
     if (fit.roi.median.xTofts$convergence != 0) {
     cat("A problem occured when fitting the Tofts model to median intensity/concentration data across the ROI (convergence code not equal to zero). Optimization algorithm did not converge. This may occur when a large fraction of voxels within the ROI are non-enhancing, as these are not excluded from analysis of the median intensity/concentration profile",
     "\n")
     }
     }
     roi.model <- roi.modelT
     if (method.optimization == "L-BFGS-B")
     fit.roi.median.Tofts <- try(optim(p0.T.median,
     Obj_roi, model = roi.model, t = map.times,
     dt = diff(map.times), cp = aif, roi = roi,
     method = "L-BFGS-B", lower = lower.wholeT,
     upper = upper.wholeT, hessian = TRUE), silent = try.silent)
     if (method.optimization != "L-BFGS-B")
     fit.roi.median.Tofts <- try(optim(p0.T.median,
     Obj_roi, model = roi.model, t = map.times,
     dt = diff(map.times), cp = aif, roi = roi,
     method = method.optimization, hessian = TRUE),
     silent = try.silent)
     if (class(fit.roi.median.Tofts) != "try-error") {
     if (fit.roi.median.Tofts$convergence == 0) {
     roi.median.fitted.Tofts <- roi.model(p = fit.roi.median.Tofts$par,
     t = map.times, dt = diff(map.times), cp = aif)
     if (tlag.Tofts.on == FALSE) {
     param.est.whole.roi.Tofts <- list(fit.roi.median.Tofts$par[1],
     fit.roi.median.Tofts$par[2])
     names(param.est.whole.roi.Tofts) <- c("Ktrans",
     "kep")
     }
     if (tlag.Tofts.on == TRUE) {
     param.est.whole.roi.Tofts <- list(fit.roi.median.Tofts$par[1],
     fit.roi.median.Tofts$par[2], fit.roi.median.Tofts$par[3])
     names(param.est.whole.roi.Tofts) <- c("Ktrans",
     "kep", "tlag")
     }
     }
     }
     if (class(fit.roi.median.Tofts) == "try-error") {
     cat("A problem occured when fitting the Tofts model to median intensity/concentration data across the ROI (a try-error occured). Optimization algorithm did not converge. This may occur when a large fraction of voxels within the ROI are non-enhancing, as these are not excluded from analysis of the median intensity/concentration profile",
     "\n")
     }
     if (class(fit.roi.median.Tofts) != "try-error") {
     if (fit.roi.median.Tofts$convergence != 0) {
     cat("A problem occured when fitting the Tofts model to median intensity/concentration data across the ROI (convergence code not equal to zero). Optimization algorithm did not converge. This may occur when a large fraction of voxels within the ROI are non-enhancing, as these are not excluded from analysis of the median intensity/concentration profile",
     "\n")
     }
     }
     roi.model <- roi.modelxT
     IRF.out <- calchFUNC(vector.times = map.times,
     AIF = aif, map_cc_slice = cc.median, vp.nom = param.est.whole.roi.xTofts$vb,
     kep.nom = param.est.whole.roi.xTofts$kep)
     p0.T <- c(IRF.out$AUChMRTh, IRF.out$AUChMRTh/IRF.out$AUCh)
     names(p0.T) <- c("Ktrans", "kep")
     if (est.per.voxel.tlag == FALSE) {
     p0.xT <- c(IRF.out$AUChMRTh, IRF.out$AUChMRTh/IRF.out$AUCh,
     param.est.whole.roi.xTofts$vb)
     names(p0.xT) <- c("Ktrans", "kep", "vb")
     }
     if (est.per.voxel.tlag == TRUE) {
     p0.xT <- c(IRF.out$AUChMRTh, IRF.out$AUChMRTh/IRF.out$AUCh,
     param.est.whole.roi.xTofts$vb, 0.05)
     names(p0.xT) <- c("Ktrans", "kep", "vb", "tlag")
     }
     AUCcorr <- IRF.out$AUCh
     AUCMRTcorr <- IRF.out$AUChMRTh
     AUCMRTcorr.divby.AUCcorr <- IRF.out$AUChMRTh/IRF.out$AUCh
     IRF.results <- c(AUCcorrnom, AUCMRTcorrnom, AUCMRTcorrnom.divby.AUCcorrnom,
     AUCcorr, AUCMRTcorr, AUCMRTcorr.divby.AUCcorr)
     names(IRF.results) <- c("AUCcorrnom(ve)", "AUCMRTcorrnom(Ktrans)",
     "AUCMRTcorrnom.divby.AUCcorrnom(kep)", "AUCcorr(ve)",
     "AUCMRTcorr(Ktrans)", "AUCMRTcorr.divby.AUCcorr(kep)")
     cat("..done in", format((proc.time()[3] - ptm)/60,
     digits = 2), "minutes.", "\n")
     cat("--------", "\n")
     roi.model <- roi.modelxT
     if (class(fit.roi.median.xTofts) != "try-error") {
     if (fit.roi.median.xTofts$convergence == 0) {
     RSS <- sum((roi.model(p = fit.roi.median.xTofts$par,
     t = map.times, dt = diff(map.times), cp = aif) -
     roi)^2)
     param_est <- fit.roi.median.xTofts$par
     nD <- nt
     nP <- length(param_est)
     hess <- fit.roi.median.xTofts$hessian
     cov <- try((nP * RSS/(nD - nP)) * solve(hess),
     silent = try.silent)
     cv <- param_est
     cv[] <- NA
     if (class(cov) != "try-error") {
     sd <- try(sqrt(diag(cov)), silent = try.silent)
     if (class(sd) != "try-error") {
     cv <- sd/param_est * 100
     cv.whole.roi.xTofts <- as.numeric(format(cv,
     digits = 1))
     }
     }
     param.roi <- as.numeric(format(param_est,
     digits = 3))
     }
     }
     if (AIF.shift == "ARTERY" | AIF.shift == "VEIN")
     aif.shifted <- aif.shift.func(map.times, aif,
     param.est.whole.roi.xTofts$tlag)
     if (show.rt.fits == TRUE) {
     if (AIF.shift == "ARTERY") {
     dev.new(width = 6, height = 6, xpos = 1500,
     ypos = 0)
     plot(map.times, aif, ylim = c(0, max(aif,
     aif.shifted)), xlab = "min", ylab = "contrast agent",
     main = paste("Shifted Arterial Input Function (",
     format(param.est.whole.roi.xTofts$tlag *
     60, digits = 3), " sec)", sep = ""),
     type = "n")
     lines(map.times, aif, col = "red")
     lines(map.times, aif.shifted)
     legend(x = 0.55 * max(map.times), y = max(aif,
     aif.shifted), c("raw AIF", "shifted AIF"),
     c("red", "black"))
     }
     if (AIF.shift == "VEIN") {
     dev.new(width = 6, height = 6, xpos = 1500,
     ypos = 0)
     plot(map.times, aif, ylim = c(0, max(aif,
     aif.shifted)), xlab = "min", ylab = "contrast agent",
     main = paste("Shifted Venous Input Function (",
     format(param.est.whole.roi.xTofts$tlag *
     60, digits = 3), " sec)", sep = ""),
     type = "n")
     lines(map.times, aif, col = "blue")
     lines(map.times, aif.shifted)
     legend(x = 0.55 * max(map.times), y = max(aif,
     aif.shifted), c("raw VIF", "shifted VIF"),
     c("blue", "black"))
     }
     if (AIF.shift == "NONE") {
     dev.new(width = 6, height = 6, xpos = 1500,
     ypos = 0)
     plot(map.times, aif, ylim = c(0, max(aif)),
     xlab = "min", ylab = "contrast agent",
     main = "Vascular Input Function", type = "n")
     lines(map.times, aif)
     legend(x = 0.55 * max(map.times), y = max(aif),
     c("raw VIF"), c("black"))
     }
     }
     if (show.rt.fits == TRUE) {
     dev.new(width = 6, height = 6, xpos = 1500,
     ypos = 575)
     plot(map.times, roi, xlab = "min", ylab = "contrast agent",
     main = "median contrast agent conc; Tofts=blue, xTofts=red",
     cex = 2)
     text(x = 0.65 * max(map.times), y = 0.4 * max(roi,
     na.rm = TRUE), labels = "EXTENDED TOFTS PARAMS",
     cex = 1)
     if (class(fit.roi.median.xTofts) != "try-error") {
     if (fit.roi.median.xTofts$convergence ==
     0) {
     text(x = 0.65 * max(map.times), y = 0.35 *
     max(roi, na.rm = TRUE), labels = paste(expression(K^trans),
     " = ", format(param.est.whole.roi.xTofts$Ktrans,
     digits = 3), " 1/min (", cv.whole.roi.xTofts[1],
     "%)", sep = ""), cex = 1)
     text(x = 0.65 * max(map.times), y = 0.3 *
     max(roi, na.rm = TRUE), labels = paste(expression(k_ep),
     " = ", format(param.est.whole.roi.xTofts$kep,
     digits = 3), " 1/min (", cv.whole.roi.xTofts[2],
     "%)", sep = ""), cex = 1)
     text(x = 0.65 * max(map.times), y = 0.25 *
     max(roi, na.rm = TRUE), labels = paste(expression(v_b),
     " = ", format(param.est.whole.roi.xTofts$vb,
     digits = 3), " dimensionless (", cv.whole.roi.xTofts[3],
     "%)", sep = ""), cex = 1)
     if (AIF.shift != "NONE")
     text(x = 0.65 * max(map.times), y = 0.2 *
     max(roi, na.rm = TRUE), labels = paste(expression(t_lag),
     " = ", format(60 * param.est.whole.roi.xTofts$tlag,
     digits = 3), " sec (", cv.whole.roi.xTofts[4],
     "%)", sep = ""), cex = 1)
     lines(map.times, roi.median.fitted.xTofts,
     col = "red", lwd = 5)
     }
     }
     if (class(fit.roi.median.Tofts) != "try-error") {
     if (fit.roi.median.Tofts$convergence == 0) {
     lines(map.times, roi.median.fitted.Tofts,
     col = "blue", lwd = 2)
     }
     }
     }
     }
     if (file.original == FALSE) {
     fix.tlag <- TRUE
     if (file.format == "matlab") {
     mat_data <- readMat(file)
     args <- mat_data$args[, , 1]
     results_file <- args$resultsfile
     ID.visit <- args$IDvisit
     slice <- args$slice
     method.optimization <- args$methodoptimization
     show.rt.fits <- args$showrtfits
     param.for.avdt <- param.for.avdt
     range.map <- args$rangemap
     cutoff.map <- args$cutoffmap
     lo <- args$lo
     est.per.voxel.tlag <- args$estpervoxeltlag
     p0.xT <- mat_data$p0xT
     p0.T <- mat_data$p0T
     AIF.shift <- args$AIFshift
     nx <- mat_data$nx
     ny <- mat_data$ny
     nt <- mat_data$nt
     map_cc_slice <- mat_data$cc
     map_cc_roi <- mat_data$ccroi
     map.times <- mat_data$maptimes
     aif <- mat_data$aif
     aif.shifted <- mat_data$aifshifted
     map.KtransxT <- mat_data$mapKtransxT
     map.tlagxT <- mat_data$maptlagxT
     map.kepxT <- mat_data$mapkepxT
     map.vexT <- mat_data$mapvexT
     map.vbxT <- mat_data$mapvbxT
     map.KtransT.cv <- mat_data$mapKtransTcv
     map.kepT.cv <- mat_data$mapkepTcv
     map.KtransxT.cv <- mat_data$mapKtransxTcv
     map.tlagxT.cv <- mat_data$maptlagxTcv
     map.kepxT.cv <- mat_data$mapkepxTcv
     map.vbxT.cv <- mat_data$mapvbxTcv
     map.fitfailuresxT <- mat_data$mapfitfailuresxT
     map.KtransT <- mat_data$mapKtransT
     map.kepT <- mat_data$mapkepT
     map.veT <- mat_data$mapveT
     map.fitfailuresT <- mat_data$mapfitfailuresT
     mask.roi <- mat_data$maskroi
     param.est.medianT <- mat_data$paramestmedianT
     param.est.medianxT <- mat_data$paramestmedianxT
     cc.median <- mat_data$ccmedian
     roi.median.fitted <- mat_data$roimedianfitted
     param.est.whole.roi <- mat_data$paramestwholeroi
     map.AIC.compare <- mat_data$mapAICcompare
     map.AIC.T <- mat_data$mapAICT
     map.EF <- mat_data$mapEF
     map.AIC.xT <- mat_data$mapAICxT
     roi.median.fitted.Tofts <- mat_data$roimedianfittedTofts
     param.est.whole.roi.Tofts <- mat_data$paramestwholeroiTofts
     roi.median.fitted.xTofts <- mat_data$roimedianfittedxTofts
     param.est.whole.roi.xTofts <- mat_data$paramestwholeroixTofts
     cv.whole.roi.xTofts <- mat_data$cvwholeroixTofts
     params.xTofts <- mat_data$paramsxTofts
     rm(mat_data)
     }
     if (file.format == "RData") {
     load(file)
     args <- dcemri.data$args
     results_file <- args$resultsfile
     ID.visit <- args$IDvisit
     slice <- args$slice
     method.optimization <- args$methodoptimization
     show.rt.fits <- args$showrtfits
     param.for.avdt <- param.for.avdt
     range.map <- args$rangemap
     cutoff.map <- args$cutoffmap
     lo <- args$lo
     p0.xT <- dcemri.data$p0xT
     p0.T <- dcemri.data$p0T
     AIF.shift <- args$AIFshift
     est.per.voxel.tlag <- args$estpervoxeltlag
     nx <- dcemri.data$nx
     ny <- dcemri.data$ny
     nt <- dcemri.data$nt
     export.matlab <- args$exportmatlab
     map_cc_slice <- dcemri.data$cc
     map_cc_roi <- dcemri.data$ccroi
     map.times <- dcemri.data$maptimes
     aif <- dcemri.data$aif
     aif.shifted <- dcemri.data$aifshifted
     map.KtransT.cv <- dcemri.data$mapKtransTcv
     map.kepT.cv <- dcemri.data$mapkepTcv
     map.KtransxT.cv <- dcemri.data$mapKtransxTcv
     map.tlagxT.cv <- dcemri.data$maptlagxTcv
     map.kepxT.cv <- dcemri.data$mapkepxTcv
     map.vbxT.cv <- dcemri.data$mapvbxTcv
     map.KtransxT <- dcemri.data$mapKtransxT
     map.tlagxT <- dcemri.data$maptlagxT
     map.kepxT <- dcemri.data$mapkepxT
     map.vexT <- dcemri.data$mapvexT
     map.vbxT <- dcemri.data$mapvbxT
     map.fitfailuresxT <- dcemri.data$mapfitfailuresxT
     map.KtransT <- dcemri.data$mapKtransT
     map.kepT <- dcemri.data$mapkepT
     map.veT <- dcemri.data$mapveT
     map.fitfailuresT <- dcemri.data$mapfitfailuresT
     mask.roi <- dcemri.data$maskroi
     param.est.medianT <- dcemri.data$paramestmedianT
     param.est.medianxT <- dcemri.data$paramestmedianxT
     cc.median <- dcemri.data$ccmedian
     roi.median.fitted <- dcemri.data$roimedianfitted
     param.est.whole.roi <- dcemri.data$paramestwholeroi
     map.AIC.compare <- dcemri.data$mapAICcompare
     map.AIC.T <- dcemri.data$mapAICT
     map.EF <- dcemri.data$mapEF
     map.AIC.xT <- dcemri.data$mapAICxT
     roi.median.fitted.Tofts <- dcemri.data$roimedianfittedTofts
     param.est.whole.roi.Tofts <- dcemri.data$paramestwholeroiTofts
     roi.median.fitted.xTofts <- dcemri.data$roimedianfittedxTofts
     param.est.whole.roi.xTofts <- dcemri.data$paramestwholeroixTofts
     cv.whole.roi.xTofts <- dcemri.data$cvwholeroixTofts
     params.xTofts <- dcemri.data$paramsxTofts
     rm(dcemri.data)
     }
     modeltype1 <- "xTofts"
     modeltype2 <- "Tofts"
     }
     if (file.original == TRUE) {
     cat("fitting", modeltype1, "and", modeltype2,
     "models to ROI voxels...", "\n")
     ptm <- proc.time()[3]
     t <- map.times
     map.KtransxT <- matrix(NA, nrow = nx, ncol = ny)
     map.kepxT <- matrix(NA, nrow = nx, ncol = ny)
     map.vexT <- matrix(NA, nrow = nx, ncol = ny)
     map.vbxT <- matrix(NA, nrow = nx, ncol = ny)
     map.tlagxT <- matrix(NA, nrow = nx, ncol = ny)
     map.KtransxT.cv <- matrix(NA, nrow = nx, ncol = ny)
     map.kepxT.cv <- matrix(NA, nrow = nx, ncol = ny)
     map.vbxT.cv <- matrix(NA, nrow = nx, ncol = ny)
     map.tlagxT.cv <- matrix(NA, nrow = nx, ncol = ny)
     map.fitfailuresxT <- matrix(NA, nrow = nx, ncol = ny)
     map.OptimValuexT <- matrix(NA, nrow = nx, ncol = ny)
     map.KtransT <- matrix(NA, nrow = nx, ncol = ny)
     map.kepT <- matrix(NA, nrow = nx, ncol = ny)
     map.veT <- matrix(NA, nrow = nx, ncol = ny)
     map.KtransT.cv <- matrix(NA, nrow = nx, ncol = ny)
     map.kepT.cv <- matrix(NA, nrow = nx, ncol = ny)
     map.fitfailuresT <- matrix(NA, nrow = nx, ncol = ny)
     map.OptimValueT <- matrix(NA, nrow = nx, ncol = ny)
     map.AIC.compare <- matrix(0, nrow = nx, ncol = ny)
     map.AIC.T <- matrix(NA, nrow = nx, ncol = ny)
     map.EF <- matrix(NA, nrow = nx, ncol = ny)
     map.AIC.xT <- matrix(NA, nrow = nx, ncol = ny)
     cc_fittedxT <- array(0, dim = c(nx, ny, nt))
     cc_fittedT <- array(0, dim = c(nx, ny, nt))
     nv <- 1
     nv1_q <- trunc(quantile(1:length(mask.roi[mask.roi ==
     1]), probs = c(0.1, 0.2, 0.3, 0.4, 0.5, 0.6,
     0.7, 0.8, 0.9, 1)))
     if (show.rt.fits == TRUE)
     dev.new(xpos = 3500, ypos = 0)
     ptm_slice <- proc.time()[3]
     GTzero <- function(x) {
     length(as.vector(x > 0)[as.vector(x > 0) ==
     TRUE])/length(x)
     }
     for (x in 1:nx) {
     for (y in 1:ny) {
     if (mask.roi[x, y] == 1 & GTzero(map_cc_slice[x,
     y, ]) <= fracGTzero) {
     map.fitfailuresxT[x, y] <- -2
     map.fitfailuresT[x, y] <- -2
     }
     if (mask.roi[x, y] == 1 & GTzero(map_cc_slice[x,
     y, ]) > fracGTzero)
     map.EF[x, y] <- 1
     if (mask.roi[x, y] == 1 & GTzero(map_cc_slice[x,
     y, ]) > fracGTzero) {
     nv <- nv + 1
     roi <- map_cc_slice[x, y, ]
     fix.tlag <- TRUE
     roi.model <- roi.modelxT
     if (verbose == TRUE) {
     cat("x =", x, "\n")
     cat("y =", y, "\n")
     cat("contrast agent curve =", roi, "\n")
     cat("fitting xTofts model to voxel data...")
     }
     if (method.optimization == "L-BFGS-B")
     fit_roi <- try(optim(p0.xT, Obj_roi,
     model = roi.model, t = map.times, dt = diff(map.times),
     cp = aif, roi = roi, method = "L-BFGS-B",
     lower = c(lo, lo, lo), upper = c(Inf,
     Inf, Inf), hessian = TRUE), silent = try.silent)
     if (method.optimization != "L-BFGS-B")
     fit_roi <- try(optim(p0.xT, Obj_roi,
     model = roi.model, t = map.times, dt = diff(map.times),
     cp = aif, roi = roi, method = method.optimization,
     hessian = TRUE), silent = try.silent)
     if (verbose == TRUE)
     cat("done", "\n")
     if (class(fit_roi) != "try-error") {
     if (fit_roi$convergence == 0) {
     RSS <- sum((roi.model(p = fit_roi$par,
     t = map.times, dt = diff(map.times),
     cp = aif) - roi)^2)
     param_est <- fit_roi$par
     nD <- nt
     nP <- length(param_est)
     hess <- fit_roi$hessian
     cov <- try((nP * RSS/(nD - nP)) * solve(hess),
     silent = try.silent)
     if (class(cov) != "try-error") {
     sd <- try(sqrt(diag(cov)), silent = try.silent)
     if (class(sd) != "try-error") {
     cv <- sd/param_est * 100
     cv.roi.voxel <- as.numeric(format(cv,
     digits = 1))
     map.KtransxT.cv[x, y] <- cv.roi.voxel[1]
     map.kepxT.cv[x, y] <- cv.roi.voxel[2]
     map.vbxT.cv[x, y] <- cv.roi.voxel[3]
     if (est.per.voxel.tlag == TRUE)
     map.tlagxT.cv[x, y] <- cv.roi.voxel[4]
     }
     }
     map.KtransxT[x, y] <- fit_roi$par[1]
     map.kepxT[x, y] <- fit_roi$par[2]
     if (map.kepxT[x, y] < 1e-05)
     map.kepxT[x, y] <- 1e-05
     map.vexT[x, y] <- fit_roi$par[1]/map.kepxT[x,
     y]
     map.vbxT[x, y] <- fit_roi$par[3]
     if (est.per.voxel.tlag == TRUE)
     map.tlagxT[x, y] <- fit_roi$par[4]
     map.OptimValuexT[x, y] <- fit_roi$value
     }
     }
     if (class(fit_roi) == "try-error") {
     if (verbose == TRUE)
     cat("...but a <try-error> occurred",
     "\n")
     map.fitfailuresxT[x, y] <- 99
     }
     if (class(fit_roi) != "try-error")
     map.fitfailuresxT[x, y] <- fit_roi$convergence
     if (class(fit_roi) == "try-error") {
     map.KtransxT[x, y] <- NA
     map.kepxT[x, y] <- NA
     map.vexT[x, y] <- NA
     map.vbxT[x, y] <- NA
     map.tlagxT[x, y] <- NA
     }
     if (class(fit_roi) != "try-error") {
     if (fit_roi$convergence > 0) {
     map.KtransxT[x, y] <- NA
     map.kepxT[x, y] <- NA
     map.vexT[x, y] <- NA
     map.vbxT[x, y] <- NA
     map.tlagxT[x, y] <- NA
     }
     }
     if (verbose == TRUE)
     cat("simulating xTofts model at estimated parameter values...")
     if (class(fit_roi) != "try-error") {
     if (fit_roi$convergence == 0) {
     simulation <- roi.model(p = fit_roi$par,
     t = map.times, dt = diff(map.times),
     cp = aif)
     cc_fittedxT[x, y, ] <- simulation
     }
     }
     if (verbose == TRUE)
     cat("done", "\n")
     roi.model <- roi.modelT
     if (verbose == TRUE)
     cat("fitting Tofts model to voxel data...")
     if (method.optimization == "L-BFGS-B")
     fit_roiT <- try(optim(p0.T, Obj_roi,
     model = roi.model, t = map.times, dt = diff(map.times),
     cp = aif, roi = roi, method = "L-BFGS-B",
     lower = lower.wholeT, upper = upper.wholeT,
     hessian = TRUE), silent = try.silent)
     if (method.optimization != "L-BFGS-B")
     fit_roiT <- try(optim(p0.T, Obj_roi,
     model = roi.model, t = map.times, dt = diff(map.times),
     cp = aif, roi = roi, method = method.optimization,
     hessian = TRUE), silent = try.silent)
     if (verbose == TRUE)
     cat("done", "\n")
     if (class(fit_roiT) != "try-error") {
     if (fit_roiT$convergence == 0) {
     RSS <- sum((roi.model(p = fit_roiT$par,
     t = map.times, dt = diff(map.times),
     cp = aif) - roi)^2)
     nD <- nt
     nP <- length(param_est)
     param_est <- fit_roiT$par
     df <- nt - length(param_est)
     hess <- fit_roiT$hessian
     cov <- try((nP * RSS/(nD - nP)) * solve(hess),
     silent = try.silent)
     if (class(cov) != "try-error") {
     sd <- try(sqrt(diag(cov)), silent = try.silent)
     if (class(sd) != "try-error") {
     cv <- sd/param_est * 100
     cv.roi.voxel <- as.numeric(format(cv,
     digits = 1))
     map.KtransT.cv[x, y] <- cv.roi.voxel[1]
     map.kepT.cv[x, y] <- cv.roi.voxel[2]
     }
     }
     map.KtransT[x, y] <- fit_roiT$par[1]
     map.kepT[x, y] <- fit_roiT$par[2]
     if (map.kepT[x, y] < 1e-05)
     map.kepT[x, y] <- 1e-05
     map.veT[x, y] <- fit_roiT$par[1]/map.kepT[x,
     y]
     map.OptimValueT[x, y] <- fit_roiT$value
     }
     }
     if (class(fit_roiT) == "try-error") {
     map.fitfailuresT[x, y] <- 99
     if (verbose == TRUE)
     cat("...but a <try-error> occurred",
     "\n")
     }
     if (class(fit_roiT) == "try-error") {
     map.KtransT[x, y] <- NA
     map.kepT[x, y] <- NA
     map.veT[x, y] <- NA
     }
     if (class(fit_roiT) != "try-error") {
     if (fit_roiT$convergence > 0) {
     map.KtransT[x, y] <- NA
     map.kepT[x, y] <- NA
     map.veT[x, y] <- NA
     }
     }
     if (class(fit_roiT) != "try-error")
     map.fitfailuresT[x, y] <- fit_roiT$convergence
     if (verbose == TRUE)
     cat("simulating Tofts model at estimated parameter values...")
     if (class(fit_roiT) != "try-error") {
     if (fit_roiT$convergence == 0) {
     simulation_2 <- roi.model(p = fit_roiT$par,
     t = map.times, dt = diff(map.times),
     cp = aif)
     cc_fittedT[x, y, ] <- simulation_2
     }
     }
     if (verbose == TRUE)
     cat("done", "\n")
     roi.model <- roi.modelxT
     if (class(fit_roi) != "try-error" & class(fit_roiT) !=
     "try-error") {
     if (fit_roi$convergence == 0 & fit_roiT$convergence ==
     0) {
     if (verbose == TRUE)
     cat("calculating AICc values...")
     simulation_AIC <- simulation
     simulation_AIC_2 <- simulation_2
     np_1 <- length(fit_roiT$par)
     np_2 <- length(fit_roi$par) + 1
     AIC1 <- nt * log(fit_roiT$value) +
     2 * (np_1 + 1) + (2 * (np_1 + 1) *
     (np_1 + 2))/(nt - np_1 - 2)
     AIC2 <- nt * log(fit_roi$value) + 2 *
     (np_2 + 1) + (2 * (np_2 + 1) * (np_2 +
     2))/(nt - np_2 - 2)
     AIC1 <- as.numeric(format(AIC1, digits = 1))
     AIC2 <- as.numeric(format(AIC2, digits = 1))
     if (AIC2 < AIC1)
     map.AIC.compare[x, y] <- 1
     if (AIC2 >= AIC1)
     map.AIC.compare[x, y] <- 2
     map.AIC.T[x, y] <- AIC1
     map.AIC.xT[x, y] <- AIC2
     }
     }
     if (class(fit_roi) != "try-error") {
     if (fit_roi$convergence > 0) {
     map.AIC.compare[x, y] <- NA
     }
     }
     if (class(fit_roiT) != "try-error") {
     if (fit_roiT$convergence > 0) {
     map.AIC.compare[x, y] <- NA
     }
     }
     if (class(fit_roi) == "try-error" | class(fit_roiT) ==
     "try-error")
     map.AIC.compare[x, y] <- NA
     if (verbose == TRUE) {
     cat("done", "\n")
     cat("======================", "\n")
     }
     if (show.rt.fits == TRUE) {
     if (class(fit_roiT) != "try-error" &
     class(fit_roi) != "try-error") {
     if (fit_roiT$convergence == 0 & fit_roi$convergence ==
     0) {
     plot(map.times, roi, ylab = "contrast agent",
     xlab = "min", main = paste(modeltype1,
     "(red) and", modeltype2, "(blue)"),
     cex = 3)
     lines(map.times, simulation, col = "red",
     lwd = 5)
     lines(map.times, simulation_2, col = "blue",
     lwd = 2)
     }
     }
     }
     if (nv == 2)
     cat("progress: ")
     if (nv == nv1_q[[1]])
     cat("10%..")
     if (nv == nv1_q[[2]])
     cat("20%..")
     if (nv == nv1_q[[3]])
     cat("30%..")
     if (nv == nv1_q[[4]])
     cat("40%..")
     if (nv == nv1_q[[5]])
     cat("50%..")
     if (nv == nv1_q[[6]])
     cat("60%..")
     if (nv == nv1_q[[7]])
     cat("70%..")
     if (nv == nv1_q[[8]])
     cat("80%..")
     if (nv == nv1_q[[9]])
     cat("90%..", "\n")
     if (nv == nv1_q[[10]] - 10)
     cat("..10")
     if (nv == nv1_q[[10]] - 9)
     cat("..9..")
     if (nv == nv1_q[[10]] - 8)
     cat("8..")
     if (nv == nv1_q[[10]] - 7)
     cat("7..")
     if (nv == nv1_q[[10]] - 6)
     cat("6..")
     if (nv == nv1_q[[10]] - 5)
     cat("5..")
     if (nv == nv1_q[[10]] - 4)
     cat("4..")
     if (nv == nv1_q[[10]] - 3)
     cat("3..")
     if (nv == nv1_q[[10]] - 2)
     cat("2..")
     if (nv == nv1_q[[10]] - 1)
     cat("1..", "\n")
     }
     }
     }
     cat("..done in", format((proc.time()[3] - ptm)/60,
     digits = 2), "minutes.", "\n")
     cat("--------", "\n")
     graphics.off()
     KtransxT.median <- median(map.KtransxT[map.KtransxT >
     0 & map.fitfailuresxT == 0], na.rm = TRUE)
     kepxT.median <- median(map.kepxT[map.kepxT >
     0 & map.fitfailuresxT == 0], na.rm = TRUE)
     vexT.median <- median(map.vexT[map.vexT > 0 &
     map.fitfailuresxT == 0], na.rm = TRUE)
     vbxT.median <- median(map.vbxT[map.vbxT >= 0 &
     map.fitfailuresxT == 0], na.rm = TRUE)
     if (est.per.voxel.tlag == TRUE)
     tlagxT.median <- median(map.tlagxT[map.tlagxT >=
     0 & map.fitfailuresxT == 0], na.rm = TRUE)
     if (est.per.voxel.tlag == FALSE)
     tlagxT.median <- NA
     fitfailuresxT.total <- length(map.fitfailuresxT[map.fitfailuresxT >=
     1])/length(map.fitfailuresxT[map.fitfailuresxT >=
     0]) * 100
     param.est.medianxT <- list(KtransxT.median, kepxT.median,
     vexT.median, vbxT.median, tlagxT.median, fitfailuresxT.total)
     names(param.est.medianxT) <- c("Ktrans.median",
     "kep.median", "ve.median", "vb.median", "tlag.median",
     "percent.fitfailures")
     KtransT.median <- median(map.KtransT[map.KtransT >
     0 & map.fitfailuresT == 0], na.rm = TRUE)
     kepT.median <- median(map.kepT[map.kepT > 0 &
     map.fitfailuresT == 0], na.rm = TRUE)
     veT.median <- median(map.veT[map.veT > 0 & map.fitfailuresT ==
     0], na.rm = TRUE)
     fitfailuresT.total <- length(map.fitfailuresT[map.fitfailuresT >=
     1])/length(map.fitfailuresT[map.fitfailuresT >=
     0]) * 100
     param.est.medianT <- list(KtransT.median, kepT.median,
     veT.median, fitfailuresT.total)
     names(param.est.medianT) <- c("Ktrans.median",
     "kep.median", "ve.median", "percent.fitfailures")
     }
     if (file.original == TRUE) {
     if (file == "concatenate.KAT.with.KAT.checkData.RData")
     ID.visit <- paste(strsplit(results_file, split = ".mat")[[1]],
     "_s", slice, sep = "")
     if (file != "concatenate.KAT.with.KAT.checkData.RData")
     ID.visit <- paste(strsplit(file, split = ".mat")[[1]],
     "_s", slice, sep = "")
     }
     if (file.original == FALSE)
     ID.visit <- strsplit(file, split = ".mat")[[1]]
     ID.visit <- strsplit(ID.visit, split = "/")
     ID.visit <- ID.visit[[1]][length(ID.visit[[1]])]
     IDvp <- strsplit(ID.visit, split = "_")
     ID.visit.forplot <- paste(IDvp[[1]][1], ".", IDvp[[1]][2],
     ".", IDvp[[1]][3], ".", IDvp[[1]][4], sep = "")
     DATE <- date()
     if (length(strsplit(DATE, split = " ")[[1]]) ==
     2) {
     full_date <- strsplit(DATE, split = " ")[[1]]
     full_date_1 <- full_date[1]
     full_date_1 <- strsplit(full_date_1, split = " ")[[1]]
     full_date_2 <- full_date[2]
     full_date_2 <- strsplit(full_date_2, split = " ")[[1]]
     month <- full_date_1[2]
     day <- full_date_2[1]
     time <- full_date_2[2]
     year <- full_date_2[3]
     }
     if (length(strsplit(DATE, split = " ")[[1]]) ==
     1) {
     full_date <- strsplit(DATE, split = " ")[[1]]
     month <- full_date[2]
     day <- full_date[3]
     time <- full_date[4]
     year <- full_date[5]
     }
     year <- strsplit(year, split = "")[[1]]
     year <- paste(year[3], year[4], sep = "")
     time_concat <- strsplit(time, split = ":")[[1]]
     time_concat <- paste(paste(time_concat[1], time_concat[2],
     sep = ""), time_concat[3], sep = "")
     DATE <- paste(paste(paste(paste(day, month, sep = ""),
     year, sep = ""), "-", sep = ""), time_concat,
     sep = "")
     filename3 <- paste(ID.visit, "_KAT_", DATE, ".mat",
     sep = "")
     filename3 <- sub(".RData", "", filename3)
     if (file.original == FALSE) {
     roi.model <- roi.modelxT
     }
     if (param.for.avdt == "Ktrans")
     MAP <- map.KtransxT
     if (param.for.avdt == "kep")
     MAP <- map.kepxT
     if (param.for.avdt == "ve")
     MAP <- map.vexT
     if (param.for.avdt == "vb")
     MAP <- map.vbxT
     if (param.for.avdt == "tlag")
     MAP <- map.tlagxT
     x_min <- 0
     x_max <- nx
     y_min <- 0
     y_max <- ny
     for (xx in 1:nx) {
     if (sum(MAP[xx, ], na.rm = TRUE) > 0) {
     x_min <- xx - 3
     break
     }
     if (sum(MAP[xx, ], na.rm = TRUE) == 0)
     xx <- xx + 1
     }
     for (y in 1:ny) {
     if (sum(MAP[, y], na.rm = TRUE) > 0) {
     y_min <- y - 3
     break
     }
     if (sum(MAP[, y], na.rm = TRUE) == 0)
     y <- y + 1
     }
     for (xx in nx:0) {
     if (sum(MAP[xx, ], na.rm = TRUE) > 0) {
     x_max <- xx + 3
     break
     }
     if (sum(MAP[xx, ], na.rm = TRUE) == 0)
     xx <- xx - 1
     }
     for (y in ny:0) {
     if (sum(MAP[, y], na.rm = TRUE) > 0) {
     y_max <- y + 3
     break
     }
     if (sum(MAP[, y], na.rm = TRUE) == 0)
     y <- y - 1
     }
     MAP_ul_s1 <- MAP[MAP > 0]
     MAP_ul_s1 <- sort(MAP_ul_s1)
     MAP_ul <- range.map * (max(MAP_ul_s1[1:length(MAP_ul_s1) *
     cutoff.map]))
     if (file.original == FALSE) {
     if (param.for.avdt == "Ktrans")
     MAP <- map.KtransxT
     if (param.for.avdt == "kep")
     MAP <- map.kepxT
     if (param.for.avdt == "ve")
     MAP <- map.vexT
     if (param.for.avdt == "vb")
     MAP <- map.vbxT
     if (param.for.avdt == "tlag")
     MAP <- map.tlagxT
     if (file.format == "matlab") {
     if (param.for.avdt == "Ktrans")
     param.median <- param.est.medianxT[, , 1]$Ktrans.median[1]
     if (param.for.avdt == "kep")
     param.median <- param.est.medianxT[, , 1]$kep.median[1]
     if (param.for.avdt == "ve")
     param.median <- param.est.medianxT[, , 1]$ve.median[1]
     if (param.for.avdt == "vb")
     param.median <- param.est.medianxT[, , 1]$vb.median[1]
     if (param.for.avdt == "tlag")
     param.median <- param.est.medianxT[, , 1]$tlag.median[1]
     }
     if (file.format == "RData") {
     if (param.for.avdt == "Ktrans")
     param.median <- param.est.medianxT$Ktrans.median
     if (param.for.avdt == "kep")
     param.median <- param.est.medianxT$kep.median
     if (param.for.avdt == "ve")
     param.median <- param.est.medianxT$ve.median
     if (param.for.avdt == "vb")
     param.median <- param.est.medianxT$vb.median
     if (param.for.avdt == "tlag")
     param.median <- param.est.medianxT$tlag.median
     }
     MAP_for_plot <- MAP
     MAP_for_plot[MAP_for_plot < 0] <- 0
     MAP_for_plot[MAP_for_plot >= MAP_ul * 0.99] <- MAP_ul *
     0.99
     dev.new(width = 6, height = 6, xpos = 1860, ypos = 578)
     plot(map.times, cc.median, xlab = "min", ylab = "contrast agent",
     main = paste(modeltype1, "(red) and", modeltype2,
     "(blue) fitted to median whole ROI data"),
     cex.main = 1, cex.axis = 1, cex.lab = 1, cex = 2)
     lines(map.times, roi.median.fitted.xTofts, col = "red",
     lwd = 5)
     lines(map.times, roi.median.fitted.Tofts, col = "blue",
     lwd = 2)
     text(x = 0.6 * max(map.times), y = 0.3 * max(cc.median,
     na.rm = TRUE), labels = "EXTENDED TOFTS PARAMS",
     cex = 1)
     text(x = 0.6 * max(map.times), y = 0.24 * max(cc.median,
     na.rm = TRUE), labels = paste(expression(K^trans),
     " = ", format(param.est.whole.roi.xTofts$Ktrans,
     digits = 3), " 1/min (", cv.whole.roi.xTofts[1],
     "%)", sep = ""), cex = 1)
     text(x = 0.6 * max(map.times), y = 0.18 * max(cc.median,
     na.rm = TRUE), labels = paste(expression(k_ep),
     " = ", format(param.est.whole.roi.xTofts$kep,
     digits = 3), " 1/min (", cv.whole.roi.xTofts[2],
     "%)", sep = ""), cex = 1)
     text(x = 0.6 * max(map.times), y = 0.12 * max(cc.median,
     na.rm = TRUE), labels = paste(expression(v_b),
     " = ", format(param.est.whole.roi.xTofts$vb,
     digits = 3), " dimensionless (", cv.whole.roi.xTofts[3],
     "%)", sep = ""), cex = 1)
     if (AIF.shift != "NONE")
     text(x = 0.6 * max(map.times), y = 0.06 * max(cc.median,
     na.rm = TRUE), labels = paste(expression(t_lag),
     " = ", format(60 * param.est.whole.roi.xTofts$tlag,
     digits = 3), " sec (", cv.whole.roi.xTofts[4],
     "%)", sep = ""), cex = 1)
     dev.new(width = 6, height = 6, xpos = 1860, ypos = 0)
     image(map.AIC.compare, xlim = c(x_min/nx, x_max/nx),
     ylim = c(y_min/ny, y_max/ny), col = palette(colorRampPalette(c("black",
     "red", "blue"))(3)), main = paste("best fit:",
     modeltype1, "(red) and", modeltype2, "(blue)"))
     image(map.AIC.compare, xlim = c(x_min/nx, x_max/nx),
     ylim = c(y_min/ny, y_max/ny), col = palette(colorRampPalette(c("black",
     "red", "blue"))(3)), main = paste("best fit:",
     modeltype1, "(red) and", modeltype2, "(blue)"))
     dev.new(width = 12.75, height = 12.75, xpos = 238,
     ypos = 0)
     image(MAP_for_plot, col = palette(colorRampPalette(c("black",
     "red", "yellow"))(1000)), xlim = c(x_min/nx,
     x_max/nx), ylim = c(y_min/ny, y_max/ny), zlim = c(0,
     MAP_ul), main = paste("Model Type=", modeltype1,
     " Median ", param.for.avdt, "=", format(param.median,
     digit = 2), " VisitID=", strsplit(args$IDvisit,
     split = "_")[[1]][1], " slice=", args$slice,
     sep = ""))
     image(MAP_for_plot, col = palette(colorRampPalette(c("black",
     "red", "yellow"))(1000)), xlim = c(x_min/nx,
     x_max/nx), ylim = c(y_min/ny, y_max/ny), zlim = c(0,
     MAP_ul), main = paste("Model Type=", modeltype1,
     " Median ", param.for.avdt, "=", format(param.median,
     digit = 2), " VisitID=", strsplit(args$IDvisit,
     split = "_")[[1]][1], " slice=", args$slice,
     sep = ""))
     text(x_min/nx + (x_max/nx - x_min/nx) * 0.97,
     y_min/ny + (y_max/ny - y_min/ny) * 0.99, "close",
     col = "green")
     text(x_min/nx + (x_max/nx - x_min/nx) * 0.05,
     y_min/ny + (y_max/ny - y_min/ny) * 0.99, "print to PDF",
     col = "green")
     text(x_min/nx + (x_max/nx - x_min/nx) * 0.84,
     y_min/ny + (y_max/ny - y_min/ny) * 0.01, paste("KAT for DCEMRI v",
     KAT.version, ", Genentech PTPK", sep = ""),
     col = "darkgrey")
     legend <- seq(0, MAP_ul, by = 0.001)
     dim(legend) <- c(1, length(legend))
     dev.new(width = 2.5, height = 12.75, xpos = 0,
     ypos = 0)
     image(legend, col = palette(colorRampPalette(c("black",
     "red", "yellow"))(1000)), zlim = c(0, MAP_ul),
     xaxt = "n", yaxt = "n", xlab = "0", main = format(MAP_ul,
     digits = 3), ylab = expression(min^-1), cex.lab = 1.25,
     cex.main = 1.05)
     image(legend, col = palette(colorRampPalette(c("black",
     "red", "yellow"))(1000)), zlim = c(0, MAP_ul),
     xaxt = "n", yaxt = "n", xlab = "0", main = format(MAP_ul,
     digits = 3), ylab = expression(min^-1), cex.lab = 1.25,
     cex.main = 1.05)
     if (AIF.shift == "ARTERY") {
     dev.new(width = 5.15, height = 4.4, xpos = 1500,
     ypos = 0)
     plot(map.times, aif, ylim = c(0, max(aif, aif.shifted)),
     xlab = "min", ylab = "contrast agent", main = paste("Shifted Arterial Input Function (",
     format(param.est.whole.roi.xTofts$tlag *
     60, digits = 3), " sec)", sep = ""),
     type = "n")
     lines(map.times, aif, col = "red")
     lines(map.times, aif.shifted)
     legend(x = 0.55 * max(map.times), y = max(aif,
     aif.shifted), c("raw AIF", "shifted AIF"),
     c("red", "black"))
     }
     if (AIF.shift == "VEIN") {
     dev.new(width = 5.15, height = 4.4, xpos = 1500,
     ypos = 0)
     plot(map.times, aif, ylim = c(0, max(aif, aif.shifted)),
     xlab = "min", ylab = "contrast agent", main = paste("Shifted Venous Input Function (",
     format(param.est.whole.roi.xTofts$tlag *
     60, digits = 3), " sec)", sep = ""),
     type = "n")
     lines(map.times, aif, col = "blue")
     lines(map.times, aif.shifted)
     legend(x = 0.55 * max(map.times), y = max(aif,
     aif.shifted), c("raw VIF", "shifted VIF"),
     c("blue", "black"))
     }
     if (AIF.shift == "NONE") {
     dev.new(width = 5.15, height = 4.4, xpos = 1500,
     ypos = 0)
     plot(map.times, aif, ylim = c(0, max(aif)),
     xlab = "min", ylab = "contrast agent", main = "Vascular Input Function",
     type = "n")
     lines(map.times, aif)
     legend(x = 0.55 * max(map.times), y = max(aif),
     c("raw VIF"), c("black"))
     }
     dev.set(4)
     inf <- 1
     newplot <- 1
     newplot2 <- 1
     legend_count <- 1
     legend_labels <- 1:1000
     legend_matrix <- matrix(0, ncol = ny, nrow = nx)
     cat("---", "\n")
     while (inf == 1) {
     z <- locator(1, type = "o", col = "green")
     xx <- round(z$x * (nx - 1) + 1)
     yy <- round(z$y * (ny - 1) + 1)
     if (legend_matrix[xx, yy] == 0) {
     legend(z$x, z$y, legend_labels[legend_count],
     col = "green", text.col = "green")
     legend_matrix[xx, yy] <- 1
     cat("Voxel Number/Coordinates: n=", legend_count,
     ", x=", xx, ", y=", yy, "\n", sep = "")
     cat("Parameter estimates (xTofts): Ktrans=",
     format(map.KtransxT[xx, yy], digits = 3),
     ", ve=", format(map.KtransxT[xx, yy]/map.kepxT[xx,
     yy], digits = 3), ", vb=", format(map.vbxT[xx,
     yy], digits = 3), ", tlag=", format(map.tlagxT[xx,
     yy], digits = 3), "\n", sep = "")
     cat("%CVs of Parameter estimates (xTofts): Ktrans=",
     format(map.KtransxT.cv[xx, yy], digits = 2),
     ", kep (Ktrans/ve)=", format(map.kepxT.cv[xx,
     yy], digits = 2), ", vb=", format(map.vbxT.cv[xx,
     yy], digits = 2), ", tlag=", format(map.tlagxT.cv[xx,
     yy], digits = 2), "\n", sep = "")
     cat("Parameter estimates (Tofts): Ktrans=",
     format(map.KtransT[xx, yy], digits = 3),
     ", ve=", format(map.KtransT[xx, yy]/map.kepT[xx,
     yy], digits = 3), "\n", sep = "")
     cat("%CVs of Parameter estimates (Tofts): Ktrans=",
     format(map.KtransT.cv[xx, yy], digits = 2),
     ", kep (Ktrans/ve)=", format(map.kepT.cv[xx,
     yy], digits = 3), "\n", sep = "")
     cat("---", "\n")
     legend_count <- legend_count + 1
     }
     xdim <- x_max - x_min
     ydim <- y_max - y_min
     if (xx > (x_max - 0.1 * xdim) && yy > (y_max -
     0.02 * ydim)) {
     graphics.off()
     dev.off()
     }
     xx_old <- xx
     yy_old <- yy
     conc <- 1:nt
     for (i in 1:nt) conc[i] <- map_cc_slice[xx,
     yy, i]
     if (xx < (x_min + 0.12 * xdim) && yy > (y_max -
     0.02 * ydim)) {
     pdf(file = paste(results_file, "-SUMMARY.pdf",
     sep = ""), height = 12, width = 15)
     layout(matrix(c(1, 2, 3, 4, 5, 6), 2, 3,
     byrow = TRUE), widths = c(1.5, 5.5, 5.5))
     par(omi = c(0.15, 0.15, 0.15, 0.15))
     image(legend, col = palette(colorRampPalette(c("black",
     "red", "yellow"))(1000)), zlim = c(0, MAP_ul),
     xaxt = "n", yaxt = "n", xlab = "0", main = format(MAP_ul,
     digits = 3), ylab = expression(min^-1),
     cex.lab = 1.5, cex.main = 1.5, cex.axis = 1.5)
     image(legend, col = palette(colorRampPalette(c("black",
     "red", "yellow"))(1000)), zlim = c(0, MAP_ul),
     xaxt = "n", yaxt = "n", xlab = "0", main = format(MAP_ul,
     digits = 3), ylab = expression(min^-1),
     cex.lab = 1.5, cex.main = 1.5, cex.axis = 1.5,
     add = TRUE)
     image(MAP_for_plot, col = palette(colorRampPalette(c("black",
     "red", "yellow"))(1000)), xlim = c(x_min/nx,
     x_max/nx), ylim = c(y_min/ny, y_max/ny),
     zlim = c(0, MAP_ul), main = paste("Model Type=",
     modeltype1, " Median ", param.for.avdt,
     "=", format(param.median, digit = 2),
     " VisitID=", strsplit(args$IDvisit,
     split = "_")[[1]][1], " slice=",
     args$slice, sep = ""), cex.axis = 1.5)
     image(MAP_for_plot, col = palette(colorRampPalette(c("black",
     "red", "yellow"))(1000)), xlim = c(x_min/nx,
     x_max/nx), ylim = c(y_min/ny, y_max/ny),
     zlim = c(0, MAP_ul), main = paste("Model Type=",
     modeltype1, " Median ", param.for.avdt,
     "=", format(param.median, digit = 2),
     " VisitID=", strsplit(args$IDvisit,
     split = "_")[[1]][1], " slice=",
     args$slice, sep = ""), cex.axis = 1.5,
     add = TRUE)
     plot(map.times, cc.median, xlab = "min",
     ylab = "contrast agent", main = "median contrast agent conc; Tofts=blue, xTofts=red",
     cex = 3, cex.axis = 1.5, cex.lab = 1.5)
     text(x = 0.6 * max(map.times), y = 0.25 *
     max(cc.median, na.rm = TRUE), labels = "EXTENDED TOFTS PARAMS",
     cex = 1.5)
     text(x = 0.6 * max(map.times), y = 0.2 *
     max(cc.median, na.rm = TRUE), labels = paste(expression(K^trans),
     " = ", format(param.est.whole.roi.xTofts$Ktrans,
     digits = 3), " 1/min (", cv.whole.roi.xTofts[1],
     "%)", sep = ""), cex = 1.5)
     text(x = 0.6 * max(map.times), y = 0.15 *
     max(cc.median, na.rm = TRUE), labels = paste(expression(v_e),
     " = ", format(param.est.whole.roi.xTofts$ve,
     digits = 3), " 1/min (", cv.whole.roi.xTofts[2],
     "%)", sep = ""), cex = 1.5)
     text(x = 0.6 * max(map.times), y = 0.1 *
     max(cc.median, na.rm = TRUE), labels = paste(expression(v_b),
     " = ", format(param.est.whole.roi.xTofts$vb,
     digits = 3), " dimensionless (", cv.whole.roi.xTofts[3],
     "%)", sep = ""), cex = 1.5)
     if (AIF.shift != "NONE")
     text(x = 0.6 * max(map.times), y = 0.05 *
     max(cc.median, na.rm = TRUE), labels = paste(expression(t_lag),
     " = ", format(60 * param.est.whole.roi.xTofts$tlag,
     digits = 3), " sec (", cv.whole.roi.xTofts[4],
     "%)", sep = ""), cex = 1.5)
     lines(map.times, roi.median.fitted.xTofts,
     col = "red", lwd = 5)
     lines(map.times, roi.median.fitted.Tofts,
     col = "blue", lwd = 2)
     text(x = 0.5 * max(map.times), 1 * max(cc.median,
     na.rm = TRUE), paste("R package version =",
     KAT.version), cex = 1.5)
     image(array(1:2, dim = c(1, 2)), col = palette(colorRampPalette(c("red",
     "blue"))(2)), zlim = c(0, 2), xaxt = "n",
     yaxt = "n", xlab = modeltype1, main = modeltype2,
     cex.lab = 1.5, cex.main = 1.5, cex.axis = 1.5)
     image(array(1:2, dim = c(1, 2)), col = palette(colorRampPalette(c("red",
     "blue"))(2)), zlim = c(0, 2), xaxt = "n",
     yaxt = "n", xlab = modeltype1, main = modeltype2,
     cex.lab = 1.5, cex.main = 1.5, cex.axis = 1.5,
     add = TRUE)
     image(map.AIC.compare, xlim = c(x_min/nx,
     x_max/nx), ylim = c(y_min/ny, y_max/ny),
     col = palette(colorRampPalette(c("black",
     "red", "blue"))(3)), main = "Summary of model discrimination analysis",
     cex.axis = 1.5, cex.lab = 1.5)
     image(map.AIC.compare, xlim = c(x_min/nx,
     x_max/nx), ylim = c(y_min/ny, y_max/ny),
     col = palette(colorRampPalette(c("black",
     "red", "blue"))(3)), main = "Summary of model discrimination analysis",
     cex.axis = 1.5, cex.lab = 1.5, add = TRUE)
     if (AIF.shift == "ARTERY") {
     plot(map.times, aif, ylim = c(0, max(aif,
     aif.shifted)), xlab = "min", ylab = "contrast agent",
     main = paste("Shifted Arterial Input Function (",
     format(param.est.whole.roi.xTofts$tlag *
     60, digits = 3), " sec)", sep = ""),
     type = "n", cex = 3, cex.axis = 1.5,
     cex.lab = 1.5)
     lines(map.times, aif, col = "red")
     lines(map.times, aif.shifted)
     legend(x = 0.55 * max(map.times), y = max(aif,
     aif.shifted), c("raw AIF", "shifted AIF"),
     c("red", "black"), cex = 2)
     }
     if (AIF.shift == "VEIN") {
     plot(map.times, aif, ylim = c(0, max(aif,
     aif.shifted)), xlab = "min", ylab = "contrast agent",
     main = paste("Shifted Venous Input Function (",
     format(param.est.whole.roi.xTofts$tlag *
     60, digits = 3), " sec)", sep = ""),
     type = "n", cex = 3, cex.axis = 1.5,
     cex.lab = 1.5)
     lines(map.times, aif, col = "blue")
     lines(map.times, aif.shifted)
     legend(x = 0.55 * max(map.times), y = max(aif,
     aif.shifted), c("raw VIF", "shifted VIF"),
     c("blue", "black"), cex = 2)
     }
     if (AIF.shift == "NONE") {
     plot(map.times, aif, ylim = c(0, max(aif)),
     xlab = "min", ylab = "contrast agent",
     main = "Vascular Input Function", type = "n",
     cex = 3, cex.axis = 1.5, cex.lab = 1.5)
     lines(map.times, aif)
     legend(x = 0.55 * max(map.times), y = max(aif),
     c("raw VIF"), c("black"), cex = 2)
     }
     dev.off()
     cat("image printed to pdf.", "\n")
     cat("---", "\n")
     }
     if (newplot == 1) {
     dev.new(width = 5.15, height = 4.4, xpos = 1500,
     ypos = 432)
     newplot <- 2
     }
     dev.set(7)
     plot(map.times, conc, xlab = "min", ylab = "contrast agent",
     ylim = c(-max(conc, na.rm = TRUE)/5, 1.4 *
     max(conc, na.rm = TRUE)), cex = 1.5, main = paste(paste("red=",
     modeltype1, ", blue=", modeltype2, " (",
     sep = ""), paste("x=", round(xx_old), ", y=",
     round(yy_old), ")", sep = ""), sep = ""))
     value_xTofts <- MAP[xx, yy]
     if (is.finite(value_xTofts) == FALSE)
     value_xTofts <- 0
     if (value_xTofts != 0) {
     if (est.per.voxel.tlag == TRUE)
     paramsxT <- c(map.KtransxT[xx, yy], map.kepxT[xx,
     yy], map.vbxT[xx, yy], map.tlagxT[xx,
     yy])
     if (est.per.voxel.tlag == FALSE)
     paramsxT <- c(map.KtransxT[xx, yy], map.kepxT[xx,
     yy], map.vbxT[xx, yy], param.est.whole.roi.xTofts$tlag)
     paramsT <- c(map.KtransT[xx, yy], map.kepT[xx,
     yy])
     roi.model <- roi.modelxT
     simulation <- roi.model(p = paramsxT, t = map.times,
     dt = diff(map.times), cp = aif)
     roi.model <- roi.modelT
     simulation_2 <- roi.model(p = paramsT, t = map.times,
     dt = diff(map.times), cp = aif)
     lines(map.times, simulation_2, col = "blue",
     lwd = 2)
     lines(map.times, simulation, col = "red",
     lwd = 2, lty = 2)
     roi.model <- roi.modelxT
     }
     text(max(map.times)/4.5, 1.4 * max(conc, na.rm = TRUE),
     "Fitted xTofts params")
     text(max(map.times)/4.5, 1.4 * max(conc, na.rm = TRUE) *
     0.93, paste(paste("Ktrans =", format(map.KtransxT[xx,
     yy], digits = 3)), "min^-1"))
     text(max(map.times)/4.5, 1.4 * max(conc, na.rm = TRUE) *
     0.86, paste("ve =", format(map.vexT[xx, yy],
     digits = 3)))
     text(max(map.times)/4.5, 1.4 * max(conc, na.rm = TRUE) *
     0.79, paste("vb =", format(map.vbxT[xx, yy],
     digits = 3)))
     if (est.per.voxel.tlag == TRUE)
     text(max(map.times)/4.5, 1.4 * max(conc,
     na.rm = TRUE) * 0.72, paste("tlag =", format(map.tlagxT[xx,
     yy], digits = 3)))
     if (newplot2 == 1) {
     dev.new(width = 5.15, height = 3, xpos = 1500,
     ypos = 864)
     newplot2 <- 2
     }
     else dev.set(8)
     if (value_xTofts != 0) {
     simulation_AIC <- simulation
     simulation_AIC_2 <- simulation_2
     if (length(simulation) == length(conc)) {
     plot(map.times, conc - simulation, xlab = "min",
     ylab = "predicated - measured", cex = 1.5,
     col = "red", main = paste("AIC(Tofts)=",
     map.AIC.T[xx, yy], " AIC(xTofts)=",
     map.AIC.xT[xx, yy], sep = ""))
     lines(locfit(conc - simulation ~ map.times,
     acri = "ici"), col = "red", lwd = 2)
     abline(h = 0, col = "black", lwd = 2)
     if (is.finite(simulation_2[1]) == TRUE) {
     points(map.times, conc - simulation_2,
     xlab = "min", cex = 1.5, col = "blue")
     lines(locfit(conc - simulation_2 ~ map.times,
     acri = "ici"), col = "blue", lwd = 2,
     lty = 2)
     }
     }
     }
     dev.set(4)
     }
     }
     if (file.original == TRUE) {
     ptm <- proc.time()[3]
     proc.time.total <- format((proc.time()[3] - ptm_total)/60,
     digits = 2)
     args <- list(as.character(file), as.character(results_file),
     as.character(method.optimization), show.rt.fits,
     as.character(param.for.avdt), range.map, cutoff.map,
     export.matlab, export.RData, verbose, show.errors,
     try.silent, fracGTzero, AIF.shift, slice, ID.visit,
     est.per.voxel.tlag)
     names(args) <- c("file", "resultsfile", "methodoptimization",
     "showrtfits", "paramforavdt", "rangemap", "cutoffmap",
     "exportmatlab", "exportRData", "verbose", "showerrors",
     "trysilent", "fracGTzero", "AIFshift", "slice",
     "IDvisit", "estpervoxeltlag")
     roiplotparams <- list(x_min, x_max, y_min, y_max,
     MAP_ul)
     names(roiplotparams) <- c("xmin", "xmax", "ymin",
     "ymax", "MAPul")
     dummy_data <- dcemri.data
     dcemri.data <- list(args, map_cc_slice, map_cc_roi,
     cc.median, map.times, aif, aif.shifted, mask.roi,
     map.KtransxT, map.KtransxT.cv, map.tlagxT,
     map.tlagxT.cv, map.kepxT, map.kepxT.cv, map.vbxT,
     map.vbxT.cv, map.vexT, map.OptimValuexT, map.fitfailuresxT,
     param.est.medianxT, roi.median.fitted.xTofts,
     param.est.whole.roi.xTofts, cv.whole.roi.xTofts,
     map.KtransT, map.KtransT.cv, map.kepT, map.kepT.cv,
     map.veT, map.OptimValueT, map.fitfailuresT,
     param.est.medianT, roi.median.fitted.Tofts,
     param.est.whole.roi.Tofts, proc.time.total,
     roiplotparams, KAT.version, map.AIC.xT, map.AIC.T,
     map.AIC.compare, nx, ny, nt, cc_fittedxT, cc_fittedT,
     p0.T, p0.xT, IRF.results, map.EF)
     names(dcemri.data) <- c("args", "cc", "ccroi",
     "ccmedian", "maptimes", "aif", "aifshifted",
     "maskroi", "mapKtransxT", "mapKtransxTcv",
     "maptlagxT", "maptlagxTcv", "mapkepxT", "mapkepxTcv",
     "mapvbxT", "mapvbxTcv", "mapvexT", "mapOptimValuexT",
     "mapfitfailuresxT", "paramestmedianxT", "roimedianfittedxTofts",
     "paramestwholeroixTofts", "cvwholeroixTofts",
     "mapKtransT", "mapKtransTcv", "mapkepT", "mapkepTcv",
     "mapveT", "mapOptimValueT", "mapfitfailuresT",
     "paramestmedianT", "roimedianfittedTofts",
     "paramestwholeroiTofts", "proctimetotal", "roiplotparams",
     "KATversion", "mapAICxT", "mapAICT", "mapAICcompare",
     "nx", "ny", "nt", "ccfittedxT", "ccfittedT",
     "p0T", "p0xT", "IRFresults", "mapEF")
     if (export.RData == TRUE) {
     cat("writing results to ", paste(results_file,
     ".RData", sep = ""), "...", sep = "", "\n")
     save(dcemri.data, file = paste(results_file,
     ".RData", sep = ""))
     }
     if (export.matlab == TRUE) {
     cat("writing results to ", paste(results_file,
     ".mat", sep = ""), "...", sep = "", "\n")
     writeMat(paste(results_file, ".mat", sep = ""),
     args = args, mapccslice = map_cc_slice, mapccroi = map_cc_roi,
     ccmedian = cc.median, maptimes = map.times,
     aif = aif, aifshifted = aif.shifted, maskroi = mask.roi,
     mapKtransxT = map.KtransxT, mapKtransxTcv = map.KtransxT.cv,
     maptlagxT = map.tlagxT, maptlagxTcv = map.tlagxT.cv,
     mapkepxT = map.kepxT, mapkepxTcv = map.kepxT.cv,
     mapvbxT = map.vbxT, mapvbxTcv = map.vbxT.cv,
     mapvexT = map.vexT, mapOptimValuexT = map.OptimValuexT,
     mapfitfailuresxT = map.fitfailuresxT, paramestmedianxT = param.est.medianxT,
     roimedianfittedxTofts = roi.median.fitted.xTofts,
     paramestwholeroixTofts = param.est.whole.roi.xTofts,
     cvwholeroixTofts = cv.whole.roi.xTofts, mapKtransT = map.KtransT,
     mapKtransTcv = map.KtransT.cv, mapkepT = map.kepT,
     mapkepTcv = map.kepT.cv, mapveT = map.veT,
     mapOptimValueT = map.OptimValueT, mapfitfailuresT = map.fitfailuresT,
     paramestmedianT = param.est.medianT, roimedianfittedTofts = roi.median.fitted.Tofts,
     paramestwholeroiTofts = param.est.whole.roi.Tofts,
     proctimetotal = proc.time.total, roiplotparams = roiplotparams,
     KATversion = KAT.version, mapAICxT = map.AIC.xT,
     mapAICT = map.AIC.T, mapAICcompare = map.AIC.compare,
     nx = nx, ny = ny, nt = nt, ccfittedxT = cc_fittedxT,
     ccfittedT = cc_fittedT, p0T = p0.T, p0xT = p0.xT,
     IRFresults = IRF.results, mapEF = map.EF)
     }
     dcemri.data <- dummy_data
     cat("..done in", format((proc.time()[3] - ptm)/60,
     digits = 2), "minutes.", "\n")
     if (export.matlab == FALSE)
     cat("Run KAT(filename.RData) to visualize results.",
     "\n")
     if (export.matlab == TRUE)
     cat("Run KAT(filename.RData) or KAT(filename.mat) to visualize results.",
     "\n")
     cat("--------", "\n")
     }
     }
     }
    }
    <bytecode: 0x3605e40>
    <environment: namespace:KATforDCEMRI>
     --- function search by body ---
    Function KAT in namespace KATforDCEMRI has this body.
     ----------- END OF FAILURE REPORT --------------
    Error in if (class(cov) != "try-error") { : the condition has length > 1
    Calls: demo ... withVisible -> eval -> eval -> runme -> system.time -> KAT
    Timing stopped at: 2.156 0.348 2.608
    Execution halted
Flavor: r-devel-linux-x86_64-fedora-gcc