Homec4science
Diffusion iCAPs public (master) Browse

iCAPs/example_normalization.m
c0ccef8aa90bmaster

Last Change
Raw File

example_normalization.m
View Options

%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
% DESCRIPTION
% this is an example script if TA has been run in subject space and data
% need to be normalized to MNI before continuing with the clustering part
% of the pipeline
%
% In this script, normalization will be applied using the SPM y*
% deformation fields
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
clc;clear all;close all;
PathData = 'example data';
Subjects = {'S1','S2'};
thresh_title = 'Alpha_5_95_Fraction_0DOT05';
%% set paths
subj_paths=fullfile(PathData,Subjects);
subj_thres_paths=fullfile(subj_paths,'TA_results','exampleToolbox','Thresholding',thresh_title);
subj_TA_paths=fullfile(subj_paths,'TA_results','exampleToolbox','TotalActivation');
subj_input_paths=fullfile(subj_paths,'TA_results','exampleToolbox','inputData');
out_thres=fullfile(subj_paths,'TA_results','exampleToolbox_MNI','Thresholding',thresh_title);
out_TA=fullfile(subj_paths,'TA_results','exampleToolbox_MNI','TotalActivation');
%% define spm deformation batch (subject-independent fields)
spm_jobman('initcfg');
matlabbatch{1}.spm.spatial.normalise.write.woptions.bb = [-78 -112 -70
78 76 85];
matlabbatch{1}.spm.spatial.normalise.write.woptions.vox = [3 3 3];
matlabbatch{1}.spm.spatial.normalise.write.woptions.prefix = 'w';
%% run deformation of SignInnov and Activity_inducing
for iS=1:length(subj_paths)
mkdir(out_thres{iS});mkdir(out_TA{iS});
% deformation field
deform_field=dir(fullfile(subj_paths{iS},'anat','Segmented','y_*'));
deform_field=fullfile(deform_field.folder,deform_field.name);
% apply to SignInnov
applyTo=dir(fullfile(subj_thres_paths{iS},'SignInnov.nii'));
applyTo=fullfile(applyTo.folder,applyTo.name);
matlabbatch{1}.spm.spatial.normalise.write.subj.def = {deform_field};
matlabbatch{1}.spm.spatial.normalise.write.subj.resample = {applyTo};
matlabbatch{1}.spm.spatial.normalise.write.woptions.interp = 4;
spm_jobman('run',matlabbatch);
movefile(fullfile(subj_thres_paths{iS},'wSignInnov.nii'),fullfile(out_thres{iS},'SignInnov.nii'))
% apply to Activity_inducing
applyTo=dir(fullfile(subj_TA_paths{iS},'Activity_inducing.nii'));
applyTo=fullfile(applyTo.folder,applyTo.name);
matlabbatch{1}.spm.spatial.normalise.write.subj.def = {deform_field};
matlabbatch{1}.spm.spatial.normalise.write.subj.resample = {applyTo};
matlabbatch{1}.spm.spatial.normalise.write.woptions.interp = 4;
spm_jobman('run',matlabbatch);
movefile(fullfile(subj_TA_paths{iS},'wActivity_inducing.nii'),fullfile(out_TA{iS},'Activity_inducing.nii'))
% apply to input mask
applyTo=dir(fullfile(subj_input_paths{iS},'mask.nii'));
applyTo=fullfile(applyTo.folder,applyTo.name);
matlabbatch{1}.spm.spatial.normalise.write.subj.def = {deform_field};
matlabbatch{1}.spm.spatial.normalise.write.subj.resample = {applyTo};
matlabbatch{1}.spm.spatial.normalise.write.woptions.interp = 1;
spm_jobman('run',matlabbatch);
movefile(fullfile(subj_input_paths{iS},'wmask.nii'),fullfile(out_TA{iS},'mask.nii'))
% apply to innovations no-nan mask
applyTo=dir(fullfile(subj_thres_paths{iS},'mask_nonan.nii'));
applyTo=fullfile(applyTo.folder,applyTo.name);
matlabbatch{1}.spm.spatial.normalise.write.subj.def = {deform_field};
matlabbatch{1}.spm.spatial.normalise.write.subj.resample = {applyTo};
matlabbatch{1}.spm.spatial.normalise.write.woptions.interp = 1;
spm_jobman('run',matlabbatch);
movefile(fullfile(subj_thres_paths{iS},'wmask_nonan.nii'),fullfile(out_thres{iS},'mask_nonan.nii'))
end
%% save a new param file for Thresholding and Acitivity inducing with updated information
for iS=1:length(subj_paths)
fHdr=cbiReadNiftiHeader(fullfile(out_thres{iS},'SignInnov.nii'));
SignInnov_4D=cbiReadNifti(fullfile(out_thres{iS},'SignInnov.nii'));
mask_nonan_3D=cbiReadNifti(fullfile(out_thres{iS},'mask_nonan.nii'));
mask_nonan_3D=~isnan(mask_nonan_3D)&mask_nonan_3D~=0;
mask_3D=cbiReadNifti(fullfile(out_TA{iS},'mask.nii'));
mask_3D=~isnan(mask_3D)&mask_3D~=0;
AI_4D=cbiReadNifti(fullfile(out_TA{iS},'Activity_inducing.nii'));
%% saving modified param and SignInnov (Thresholding)
load(fullfile(subj_thres_paths{iS},'param.mat'));
param.mask=reshape(mask_3D,[],1);
param.Dimension(1)=size(mask_3D,1);
param.Dimension(2)=size(mask_3D,2);
param.Dimension(3)=size(mask_3D,3);
param.mask_nonan=reshape(mask_nonan_3D,[],1);
% these are fields that are specific for thresholding in subject space
% and won't be required further by the pipeline
param=rmfield(param,{'PC','mask_threshold1'});
% getting 2D innovations in MNI
SignInnov=reshape(SignInnov_4D,[],size(SignInnov_4D,4));
SignInnov=SignInnov(param.mask_nonan,:)';
% masking normalized data
SignInnov_4D(~repmat(mask_nonan_3D,1,1,1,size(SignInnov_4D,4)))=nan;
% saving MNI data
save(fullfile(out_thres{iS},'param.mat'),'param','-v7.3');
save(fullfile(out_thres{iS},'SignInnov.mat'),'SignInnov','-v7.3');
hdr=cbiReadNiftiHeader(fullfile(out_thres{iS},'SignInnov.nii'));
cbiWriteNifti(fullfile(out_thres{iS},'SignInnov.nii'),SignInnov_4D,hdr,'float32');
%% saving modified param (Total Activation)
load(fullfile(subj_TA_paths{iS},'param.mat'));
param.mask=reshape(mask_3D,[],1);
param.mask_3D=mask_3D;
param.Dimension(1)=size(mask_3D,1);
param.Dimension(2)=size(mask_3D,2);
param.Dimension(3)=size(mask_3D,3);
param.IND=find(mask_3D);
param.VoxelIdx=[];
[param.VoxelIdx(:,1),param.VoxelIdx(:,2),param.VoxelIdx(:,3)]=ind2sub(size(mask_3D),param.IND);
param.NbrVoxels=length(param.IND);
% these are fields that are specific for TA in subject space
% and won't be required further by the pipeline
param=rmfield(param,{'GM_map','fHeader','weight_x','weight_y','weight_z',...
'LambdaTemp','LambdaTempFin','NoiseEstimateFin'});
% getting 2D innovations in MNI
Activity_inducing=reshape(AI_4D,[],size(AI_4D,4));
Activity_inducing=Activity_inducing(param.mask,:)';
% masking normalized data
AI_4D(~repmat(mask_3D,1,1,1,size(AI_4D,4)))=nan;
% saving MNI data
save(fullfile(out_TA{iS},'param.mat'),'param','-v7.3');
save(fullfile(out_TA{iS},'Activity_inducing.mat'),'Activity_inducing','-v7.3');
hdr=cbiReadNiftiHeader(fullfile(out_TA{iS},'Activity_inducing.nii'));
cbiWriteNifti(fullfile(out_TA{iS},'Activity_inducing.nii'),AI_4D,hdr,'float32');
end
c4science · Help