fmri.smooth {fmri}R Documentation

Smoothing Statistical Parametric Maps

Description

Perform the adaptive weights smoothing procedure

Usage

fmri.smooth(spm, hmax = 4, adaptive=TRUE,
            lkern="Gaussian", skern="Plateau", weighted=TRUE)

Arguments

spm object of class fmrispm
hmax maximum bandwidth to smooth
adaptive logical. TRUE (default) for adaptive smoothing
lkern lkern specifies the location kernel. Defaults to "Gaussian", other choices are "Triangle" and "Plateau". Note that the location kernel is applied to (x-x_j)^2/h^2, i.e. the use of "Triangle" corresponds to the Epanechnicov kernel in nonparametric kernel regression. "Plateau" specifies a kernel that is equal to 1 in the interval (0,.3), decays linearly in (.5,1) and is 0 for arguments larger than 1.
skern skern specifies the kernel for the statistical penalty. Defaults to "Plateau", the alternatives are "Triangle" and "Exp". "Plateau" specifies a kernel that is equal to 1 in the interval (0,.3), decays linearly in (.3,1) and is 0 for arguments larger than 1. lkern="Plateau" and lkern="Triangle" allow for much faster computation (saves up to 50% CPU-time). lkern="Plateau" produces a less random weighting scheme.
weighted weighted (logical) determines if weights contain the inverse of local variances as a factor (Weighted Least Squares). weighted=FALSE does not employ the heteroscedasticity of variances for the weighting scheme and is preferable if variance estimates are highly variable, e.g. for short time series.

Details

This function performs the smoothing on the Statistical Parametric Map spm.

hmax is the (maximal) bandwidth used in the last iteration. Choose adaptive as FALSE for non adaptive smoothing. lkern can be used for specifying the localization kernel. For comparison with non adaptive methods use "Gaussian" (hmax given in FWHM), for better adaptation use "Plateau" or "Triangle" (default, hmax given in voxel). For lkern="Plateau" and lkern="Triangle" thresholds may be inaccurate, due to a violation of the Gaussian random field assumption under homogeneity. lkern="Plateau" is expected to provide best results with adaptive smoothing.

skern can be used for specifying the kernel for the statistical penalty. "Plateau" is expected to provide the best results, due to a less random weighting scheme.

The function handles zero variances by assigning a large value (1e20) to these variances. Smoothing is restricted to voxel with spm$mask.

Value

object with class attributes "fmrispm" and "fmridata"

cbeta smoothed parameter estimate
var variance of the parameter
hmax maximum bandwidth used
rxyz smoothness in resel space. all directions
rxyz0 smoothness in resel space as would be achieved by a Gaussian filter with the same bandwidth. all directions
scorr array of spatial correlations with maximal lags 5, 5, 3 in x,y and z-direction.
bw vector of bandwidths (in FWHM) corresponding to the spatial correlation within the data.
dim dimension of the data cube and residuals
weights ratio of voxel dimensions
vwghts ratio of estimated variances for the stimuli given by vvector
hrf Expected BOLD response for the specified effect

Author(s)

Joerg Polzehl polzehl@wias-berlin.de, Karsten Tabelow tabelow@wias-berlin.de

References

Tabelow, K., Polzehl, J., Voss, H.U., and Spokoiny, V.. Analysing {fMRI} experiments with structure adaptive smoothing procedures, NeuroImage, 33:55-62 (2006).

Polzehl, J. and Spokoiny, V. (2006). Propagation-Separation Approach for Local Likelihood Estimation, Probab. Theory Relat. Fields 135, 335-362.

Examples

## Not run: fmri.smooth(spm, hmax = 4, lkern = "Gaussian")

[Package fmri version 1.2-6 Index]