Multimodal CT-MRI radiomics model accurately predicts lung cancer brain metastasis subtype

Multimodal brain radiomics (CT, CE-T1WI, T2WI) predicted pathological subtypes of lung cancer brain metastases with AUC 0.883. Whole-brain features outperformed tumor-only, and spatial extent was more critical than MRI sequence.

• Study of 255 patients (retrospective, training:test 7:3) used radiomics from tumor and whole-brain regions on CT, CE-T1WI, and T2WI.
• T1WI models outperformed T2WI; adding clinical data to the multimodal model raised AUC to 0.890 in the test cohort.
• Whole-brain-based models were significantly superior to tumor-region models, and spatial range had a greater impact than imaging sequence.

Automated summary

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