Hi Frank,
I hope you are well.
I'm currently running atlas-based segmentations on T1-weighted MRI NFITI files to extract the volumes of the hypothalamic regions. The atlas I'm using is by Neudorfer, who developed an atlas of the human hypothalamus:
https://pubmed.ncbi.nlm.nih.gov/32934244/
I wanted to ask you if my processing pipeline in DSI studio for the segmentations seems correct. Currently I am loading the raw NIFTI file into DSI studio (from Step 3 Fiber Tracking), then going to Regions --> Open MNI region --> atlas_labels_0.5mm.nii (which is the atlas NIFTI file from Neudorfer et al).
I believe DSI studio then automatically undergoes a normalization process, and the hypothalamus from the atlas is loaded anatomically onto the native scan, showing the segmented regions.
I then navigate to Regions --> Statistics to get the volumes of the hypothalamic regions in mm^3.
For most scans, the volume data of the hypothalamic regions aligns as expected to the literature values indicated in the Neudorfer et al paper (in Table 2). They also align anatomically to where the hypothalamus should be in the MRI scan.
I was wondering is this current strategy of opening MNI hypothalamus atlas on native data is robust via DSI Studio to extract volume data?
I've attatched a couple images of the atlas of the hypothalamus and surrounding regions laid out on the brain and it looks good on our end - just wanted to confirm the validity of the pipeline.
Thanks,
Zubina