Hi all,
I made a post yesterday as a response to an older related thread, but since my post was not linked with that thread in the archives, I thought I'd start a new thread a new more specific question. I have done model estimation on functional surface data (output by FMRIPREP pre-processing), and would like to use the surficial beta weight maps for further multivariate analyses. However I am having trouble reading the beta weight maps.
The issue is that SPM stores its gifti data in a rather unconventional way, with the actual data stored in an external file (e.g. beta_0001.dat) pointed to by the beta_0001.gii file. Realizing that the .dat file is just a binary data file, I found that I can use the matlab command:
surf_data = fread(fopen('beta_0001.dat'));
However, the data comes in at an unexpected size, a vector with length 4*n_nodes. I can read in the elements as a matrix of size [n_nodes,4], but I'm not sure what the extra columns are for. In the .gii file, it says that cdata should have dimensionality of 1, with n_nodes values.
Can someone assist with this issue, and perhaps suggest a recommended way for loading in these external binaries for gifti data?
thanks,
Nick
|