External Email - Use Caution
Dear FreeSurfer Team,
I have been analyzing brain volumes using FreeSurfer and noticed that cortical volume is consistently larger than cerebral white matter volume, which seems counterintuitive. This pattern remains even after QA, where absolute values change but the ratio stays the same. The most striking case is in the cerebellum, where cortex volume is much greater than white matter. I wanted to ask if this is an expected outcome in FreeSurfer’s segmentation or if it could indicate a systematic bias.
I was analyzing the following volumes: Cerebellar: Right-Cerebellum-White-Matter, Right-Cerebellum-Cortex, Left-Cerebellum-White-Matter, Left-Cerebellum-Cortex Global Brain: BrainSegVol, BrainSegVolNotVent, CerebralWhiteMatterVol, SubCortGrayVol, TotalGrayVol, SupraTentorialVol, SupraTentorialVolNotVent, CortexVol, rhCortexVol, lhCortexVol, lhCerebralWhiteMatterVol, rhCerebralWhiteMatterVol
As part of QA, I manually checked the segmentation and made the following corrections:
Checked skull and dura to prevent cortex invasion Reviewed sagittal and transverse sinuses to avoid misclassification Inspected orbits and middle cerebral arteries for segmentation accuracy Examined WM for extent and lesions to prevent miscalculations I have attached aseg_stats.txt for reference. I would greatly appreciate any insights on whether these volume relationships are expected.
Best regards,
Pamela González