Hi all,
I'm new to SPM and I am attempting to use it to complete a tissue segmentation on my T1 images. I'm primarily interested in extracting the volume of the lateral ventricles using the produced CSF image. Grey and white matter segmentation isn't a priority.
I'm having an issue regarding patients with significantly enlarged ventricles due to hydrocephalus. The ventricles are so large that the code is not able to successfully complete a tissue segmentation in a meaningful way. In the resulting CSF image, only a small portion of the CSF is identified. Examining the GM and WM images, the majority of the ventricle is labelled as white matter in the segmentation and then the surrounding grey and white matter are labelled as grey matter in the segmentation.
Has anyone else come across this issue previously and if so how have you handled the incorrect segmentation?
For reference, the T1 image that is warped into MNI space looks decent (albeit the image still obviously has noticeably enlarged ventricles that are much larger than the atlas image and is likely the cause of the incorrect tissue segmentation).
Best wishes,
Emily
|