High intensities in chiasmatic region seems to affect fixel analysis

Dear experts,

  1. We have high intensities in the chiasmatic region in our dMRI data. A similar problem was already described here. How can we solve that problem?

  2. This seems to also affect the fixel analysis. As shown in the attached images.

We attached raw dMRI data(1), preprocessed data(2), fod template(3) and fod with fixels(4).

image 1:

image 2:

image 3:

image 4:

Hi Ziqian,

The hyper-intensity in the DWIs is caused by a compression of signal due to EPI susceptibility distortions. Without having the requisite image data to correct said distortions, there’s likely nothing you can do about the hyper-intensity itself on a per-subject basis; at least not without grossly altering the quantitative behaviour of AFD…

The relevant fixels could be masked from the FBA analysis using a combination of a manually-defined ROI and the voxel2fixel command to produce a fixel analysis mask, and then providing this to fixelcfestats via the-mask` option.

Rob