Volumetric fMRI Searchlight

Contents

Load data

config = cosmo_config();
data_path = fullfile(config.tutorial_data_path, 'ak6', 's01');

% In this exercise, output (NIFTI files) will be written to this directory
output_data_path = config.output_data_path;

% file locations for both halves
half1_fn = fullfile(data_path, 'glm_T_stats_odd.nii');
half2_fn = fullfile(data_path, 'glm_T_stats_even.nii');
mask_fn = fullfile(data_path, 'brain_mask.nii');

% load two halves as CoSMoMVPA dataset structs.
half1_ds = cosmo_fmri_dataset(half1_fn, 'mask', mask_fn, ...
                              'targets', (1:6)', ...
                              'chunks', repmat(1, 6, 1));
half2_ds = cosmo_fmri_dataset(half2_fn, 'mask', mask_fn, ...
                              'targets', (1:6)', ...
                              'chunks', repmat(2, 6, 1));
ds = cosmo_stack({half1_ds, half2_ds});

% remove constant features (caused by liberal masking)
ds = cosmo_remove_useless_data(ds);

Define spherical neighborhood for each feature (voxel)

radius = 3; % searchlight radius in voxels

% Using cosmo_spherical_neighborhood, define a neighborhood with
% a radius of 3 voxels for each voxel. Assign the result to
% a variable named 'nbrhood'
% >@@>
nbrhood = cosmo_spherical_neighborhood(ds, 'radius', radius);
% <@@<

% Compute the number of elements in each element of nbrhood.neighbors,
% and assign the result to a variable 'roi_sizes'
% >@@>
roi_sizes = cellfun(@numel, nbrhood.neighbors);
% <@@<

% Plot a histogram of 'roi_sizes'
% >@@>
hist(roi_sizes, 100);
% <@@<

+00:00:00 [####################] -00:00:00  mean size 111.5

Run a searchlight with the cosmo_correlation_measure

% Use cosmo_searchlight with 'ds' as the dataset input,
% a function handle of cosmo_correlation_measure as the measure,
% and the spherical neighborhood just defined to run a searchlight.
% Assign the result to a variable 'ds_corr'.
% >@@>
ds_corr = cosmo_searchlight(ds, nbrhood, @cosmo_correlation_measure);
% <@@<

+00:00:01 [####################] -00:00:00  

Visualize and store the results in a NIFTI file

% Visualize the results using cosmo_plot_slices
% Hint: this function takes datasets with one sample directly as input
% >@@>
cosmo_plot_slices(ds_corr);
% <@@<

% Set output filename
output_fn = fullfile(output_data_path, ...
                     sprintf('splithalf_correlation_searchlight_r%.0f.nii', radius));

% Write output to a NIFTI file using cosmo_map2fmri
% >@@>
cosmo_map2fmri(ds_corr, output_fn);
% <@@<

Published with MATLAB® R2024a