7T susceptibility-weighted MRI vascular features plus machine learning distinguish glioma IDH status and WHO grade

Adding 7T susceptibility-weighted MRI vascular features to clinical factors yielded area under the curve 0.888 for isocitrate dehydrogenase status and 0.889 for World Health Organization grade, outperforming imaging-only models (0.816–0.843 and 0.834–0.874).

• Nine robust vascular features were retained after feature selection; models included logistic regression, support vector machine, naive Bayes, and random forest.
• Internal validation imaging-only area under the curve was 0.816–0.843 for IDH status and 0.834–0.874 for WHO grade.
• Evidence level 4, technical efficacy stage 2, retrospective single-center design without external validation.

Evidence grade: Level 4

Automated summary

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