External Email - Use Caution
Hello,
I am attempting to segment the choroid plexus using T1 and T2-FLAIR images as inputs to Samseg. On our T1 images, the choroid plexus is hardly visible, but it is much more visible on T2-FLAIR. Samseg seems to be failing to segment the choroid plexus on some images, but does a good job on others.
For the pallidum, there is a flag “--pallidum-separate” which seems to privilege the T2-FLAIR for pallidum segmentation.
Is there a similar argument or option that could be used to separate the segmentation of the choroid plexus to privilege the T2-FLAIR, such as “--choroid-separate”?
Thank you, Hudson Robb
There is not, but you could create your own. By default, it uses this table to determine which superclass each segment goes into $FREESURFER/average/samseg/20Subjects_smoothing2_down2_smoothingForAffine2/sharedGMMParameters.txt YOu can see that CP is in the CSF glass. You can try creating its own class or moving to another class If you are using the samseg script, then pass the new file in with --gmm
On 9/15/2021 2:10 PM, Robb, William Hudson wrote:
External Email - Use Caution
External Email - Use Caution
Hello,
I am attempting to segment the choroid plexus using T1 and T2-FLAIR images as inputs to Samseg. On our T1 images, the choroid plexus is hardly visible, but it is much more visible on T2-FLAIR. Samseg seems to be failing to segment the choroid plexus on some images, but does a good job on others.
For the pallidum, there is a flag “--pallidum-separate” which seems to privilege the T2-FLAIR for pallidum segmentation.
Is there a similar argument or option that could be used to separate the segmentation of the choroid plexus to privilege the T2-FLAIR, such as “--choroid-separate”?
Thank you,
Hudson Robb
Freesurfer mailing list Freesurfer@nmr.mgh.harvard.edu https://secure-web.cisco.com/1gQmXHpjt4hWJzNeQCTwUiBAAdpQTBbfGh9oHaWq5mlmiIF...
freesurfer@nmr.mgh.harvard.edu