FLAIR, ASL, and T1w sequences alone achieve highest accuracy for glioma grade, diagnosis, and IDH status in radiomics models

FLAIR alone gave 57.4% accuracy for WHO grade, ASL 81.5% for final diagnosis (glioblastoma vs astrocytoma), and T1w 66.7% for IDH status. Combining all sequences did not significantly improve performance, supporting streamlined MRI protocols.

• Analysis used 501 patients from the UCSF-PDGM dataset, extracting radiomics features from ASL, DWI, FLAIR, SWI, T1w, and T2w sequences.
• Across tasks, models consistently selected shape features and demographic variables (age, tumor sphericity).

Automated summary

RadPigeon summaries are original and for information only. They are not clinical advice.