Homec4science
Diffusion iCAPs public version 2 (master) Browse

iCAPs_v2/functions/01_TotalActivation/Generate_Innovations.m
e5f5005bb2efmaster

Last Change
Raw File

Generate_Innovations.m
View Options

%% This function generates the innovation and activity inducing signals
% from the activity related signals (output from TA)
%
% Inputs:
% - TC_OUT is a 2D matrix (n_time_points x n_ret_voxels) with the outputs
% from total activation
% - param contains the TA-relevant parameters; here, we need the fields
% 'Dimension' (size of X, Y, Z and T), 'NbrVoxels' (number of voxels
% retained for TA), 'f_Recons' (the filter that converts the
% activity-related signal into the activity-inducing one), 'f_Analyze' (the
% one with added derivation step)
%
% Outputs:
% - Innovation is a n_time_points x n_ret_voxels 2D matrix with the
% innovation signals
% - Activity_inducing is a n_time_points x n_ret_voxels 2D matrix with the
% activity-inducing signals
function [Innovation,Activity_inducing] = Generate_Innovations(TC_OUT,param)
Activity_inducing = zeros(param.Dimension(4),param.NbrVoxels);
Innovation = zeros(param.Dimension(4),param.NbrVoxels);
% Each voxel time course is deconvolved using the reconstruction filter
% (solely deconvolution)
for i=1:param.NbrVoxels
% Applies the reconstruction filter (only deconvolution)
Activity_inducing(:,i) = filter_boundary(param.filter_reconstruct.num,param.filter_reconstruct.den,TC_OUT(:,i),'normal');
% Applies the analysis filter (also encompasses the differentiation
% step)
Innovation(:,i) = filter_boundary(param.f_Analyze.num ,param.f_Analyze.den,TC_OUT(:,i),'normal');
% Those are the ways originally present in the code: I have no idea
% why they were used instead of the above option...
%Innovation(:,i) = [0;diff((Activity_inducing(:,i)))];
%Innovation(:,i) = cumsum([zeros(5,1); Innovation(6:end,i)]);
end
end
c4science · Help