Volumetric MRI segmentation

Overview

Manual 3D volumetric segmentation of tumor on MRI (typically T2/FLAIR abnormality) to derive tumor volume measurements, used as a quantitative imaging biomarker for tumor growth kinetics during active surveillance.

Used by

• PMID:37910594 — manual 3D volumetric segmentation of T2/FLAIR abnormality performed on serial MRIs (median 8 scans/patient) of 128 IDH-mutant Grade 2 gliomas under active surveillance at MSKCC, using TeraRecon iNtuition 4.4.13 with VASARI feature annotation, to compute tumor volume growth rate (TVGR) and doubling time PMID:37910594.
• PMID:28872634 — expert volumetric segmentation of tumor sub-regions (enhancing tumor [ET], non-enhancing tumor core [NET], peritumoral edema [ED]) on pre-operative multi-parametric MRI (T1, T1-Gd, T2, T2-FLAIR) for 243 glioma cases (TCGA-GBM n=135, TCGA-LGG n=108); initial labels generated by GLISTRboost then manually corrected by a board-certified neuroradiologist; >700 radiomic features extracted volumetrically via CaPTk PMID:28872634.

Notes

• Heterogeneous scan parameters across ~22 years of imaging (varying slice thickness, interslice gap, scanner type) may bias volume estimates PMID:37910594.
• Manual segmentation, although blinded and expert-reviewed, may include edema or post-surgical changes indistinguishable from tumor on MRI PMID:37910594.
• Automated/AI-assisted segmentation is called out as a prospective next step PMID:37910594.

Sources

• PMID:37910594
• PMID:28872634

This page was processed by entity-page-writer on 2026-04-15.