Hi Binod,
To be able to use PRoNTo, you need to create a nifti file for each
subject. You'll need to import your data in Matlab and convert each row
into a nifti file (see SPM routine 'nifti', ). PRoNTo has been tested
for 2D, 3D and 4D Nifti images. I cannot vouch for 1D images so it might
be something to consider when creating your images. Since you have
vertices (i.e. coordinates), you should try to build 3D images and fill
the corresponding coordinates with your data, so that weight images will
make sense anatomically speaking.
I hope this helps,
Best,
Happy 2015,
Jessica
On 12/30/2014 11:30 AM, Binod Thapa-Chhetry wrote:
> Hi all,
>
> I am working on getting freesurfer's cortical thickness surfaces as my
> predictors in pronto. I am able to extract cortical thickness values
> for each co-ordinate on the surface in a csv file where each row is a
> subject and columns are vertices. Is there any way I could use pronto
> to create kernel matrix and perform machine learning on this data.
>
> Thank you so much!
> Binod
|