Issue with Jacobian determinant maps

[ndrewvo]

Hi,

I’m trying to use the CAT12 toolbox to compute DBM maps from T1w images. I used the default parameters and saved the Jacobian determinant map. The resulting map seems to have an issue in which all voxel values are inflated (i.e., above 1), even in the background, despite these T1w images being from patients with neurodegeneration. I’ve tried running the latest 12.9 and 12.8 versions but they both result in the same issue. Only if I downgrade to 12.7 does the DBM map look like expected.

Is there a preprocessing step or setting that I'm missing? Thanks in advance for any clarification!

Best regards,
Andrew

[minoskritikos]

I had this issue too, I could only solve it by changing the registration method in the segmentation pipeline from shooting to DARTEL. I wasted a lot of time trying to figure this out, but you get correct Jacobian values using DARTEL, instead of the default shooting.

Best
Minos

[ndrewvo]

Wow, thank you so much @minoskritikos! This saved me a lot of time and headaches. Much appreciated :-)

[jcf2]

We noticed we had this issue too. Thank you @robdahn for taking a look. I wanted to point out that when you patched it, it looks like you may have referenced the wrong issue (#32 rather than this #31) on your fixes, which marked that one as closed but didn't create any followup on this one? That other issue is also the one referred to in the commit for CHANGES (137feab).

Is there any plan to create a new release anytime soon? We're looking forward to using the patched version, and it's always nice if there is a release to refer to.