NeuroGraph Utilities¶
- construct_adj(corr, threshold=5)[source]¶
create adjacency matrix from functional connectome matrix
Args:
corr (n x n numpy matrix): functional connectome matrix
Threshold (int (1- 100)): threshold for controling graph density.
the more higher the threshold, the more denser the graph. default: 5
- construct_corr(m)[source]¶
This function construct correlation matrix from the preprocessed fmri matrix Args.
m (numpy array): a preprocessed numpy matrix return: correlation matrix
- construct_data(corr, label, threshold=5)[source]¶
create pyg data object from functional connectome matrix. We use correlation as node features Args:
corr (n x n numpy matrix): functional connectome matrix
Threshold (int (1- 100)): threshold for controling graph density.
the more higher the threshold, the more denser the graph. default: 5
- parcellation(fmri, n_rois=1000)[source]¶
Prepfrom brain parcellation
Args:
fmri (numpy array): fmri image rois (int): {100, 200, 300, 400, 500, 600, 700, 800, 900, 1000}, optional, Number of regions of interest. Default=1000.
- preprocess(fmri, regs, n_rois=1000)[source]¶
Preprocess fMRI data using NeuroGraph preprocessing pipeline
Args:
fmri (numpy array): fmri image regs (numpy array): regressor array rois (int): {100, 200, 300, 400, 500, 600, 700, 800, 900, 1000}, optional, Number of regions of interest. Default=1000.