Problem with dwi2response msmt_5tt (amp2response) command for ex-vivo monkey data

[ateillac]

Dear @ThijsDhollander and @rsmith,

thanks for your time.

@ThijsDhollander

From that screenshot, I can already assure you that you don’t need the -fa option to the dhollander algorithm; the default value (of 0.2) will work just fine out of the box!

That’s what I thought by looking at the fa map but thanks for the confirmation! You can find below a screenshot of the adc map (while I’ll ask how the brain was fixed and in which fluid):

I launched the dwi2response with the dhollander algorithm and the default -fa parameter and it seems to work just fine as you can see from the voxels.mif screenshot below:

I’m really impressed by how robust this algorithm is!

@rsmith thanks a lot for the clarification of what mrtransform did to get the 5tt_regrid.mif with the linear interpolation based on the fa map. I looked at what you mentionned with the use of the option -masks in the 5ttcheck command and it did exactly what you expected (mask of the outer edge of the brain mask)! I’ll bookmark the future fix conversation and look into the strides for the data image.

Thanks a lot to all you for your reactivity, advices and explanations!

Regards,
Achille