Good afternoon,

 

I hope this email finds you well.

I am in the process of “painting” MRI surfaces from the Human connectome project in FreeSurfer, meaning creating labels on the surfaces to analyze sulcal morphology. We are interested in the spatial variability of the sulcus and thus, want to create surface spatial probability maps using FreeSurfer (please see attachment for an example of such surface probability maps). These maps would represent the likelihood of each vertex on the FreeSurfer fsaverage surface belonging to the sulcus of interest (from 0% to 100%). 

To obtain these probability maps, I:

• Painted the sulcus of interest on the surface by creating a .label file.
• Registered this .label file to fsaverage (mri_label2label)
• Converted the .label file to vol (mri_label2vol)
• Converted it to a surface file (mri_vol2surf)
• Concatenated the data to create an average (mri_concat --mean)
• Blurred the data and loaded the blurred average as an overlay onto the fsaverage pial (mris_fwhm)

Because this seems very convoluted, my question becomes: Is there a more direct way to generate these maps from the labels themselves, rather than going from label to vol, then back to surf?

Finally, I would like these maps to represent a 0-100% range of each vertex in fsaverage belonging to the sulcus of interest. However, this method creates a 0-1 range (please see colour bar in attachment). Is there a way to change this to a percentage?

Thank you very much for your time and help. Please let me know if you have any questions.

Best,
Kristi   

 

                                                                

Kristi Drudik 

Ph.D. Student, Clinical Psychology

 

Montreal Neurological Institute 

McGill University 

Montreal, Canada