MRI-based preoperative WHO grade prediction in meningioma: development and comparison of deep learning, radiomics, clinical, and combined models in multicenter data.
Authors
Affiliations (10)
Affiliations (10)
- Department of Radiology, Gazimagusa State Hospital, Famagusta, Northern Cyprus.
- Department of Neurological Surgery, School of Medicine and Public Health, University of Wisconsin, CSC K4/822, 600 Highland Avenue, Madison, WI, 53792, USA.
- Icahn School of Medicine at Mount Sinai, Biomedical Engineering and Imaging Institute, New York, NY, USA.
- Department of Neurosurgery, Istanbul Training and Research Hospital, University of Health Sciences, Istanbul, Turkey.
- Department of Neurosurgery, Bakirkoy Prof. Dr. Mazhar Osman Training and Research Hospital for Psychiatry, Neurology, and Neurosurgery, University of Health Sciences, Istanbul, Turkey.
- Department of Neurosurgery, Gazi University Faculty of Medicine, Ankara, Turkey.
- Department of Neurosurgery, Prof. Dr. Suleyman Yalcin City Hospital, Istanbul Medeniyet University, Istanbul, Turkey.
- Department of Radiology,Istanbul Training and Research Hospital, University of Health Sciences, Istanbul, Turkey.
- Department of Radiology,Bagcilar Training and Research Hospital, University of Health Sciences, Istanbul, Turkey.
- Department of Neurological Surgery, School of Medicine and Public Health, University of Wisconsin, CSC K4/822, 600 Highland Avenue, Madison, WI, 53792, USA. [email protected].
Abstract
The objective of this study was to develop and externally validate complementary magnetic resonance imaging (MRI)-based models and their stacked combinations for preoperative World Health Organization (WHO) grading of intracranial meningioma. This retrospective multicenter study included 455 histopathologically confirmed meningiomas: a four-center development cohort (n = 346) with leave-one-center-out cross-validation and an external cohort (n = 109). Four base models used contrast-enhanced T1-weighted MRI: tumor + peritumoral radiomics, an nnU-Net segmentation-based classifier, frozen DINOv3 transfer-learning features, and clinical variables. All 15 combinations were prespecified and stacked using a linear meta-learner trained on development out-of-fold predictions. Frozen models were externally evaluated without modification. WHO 2-3 tumors comprised 24.6% overall (112/455; development 20.5%; external 37.6%). External AUCs were 0.732 for nnU-Net, 0.703 for DINOv3, 0.645 for clinical, 0.628 for radiomics. The multimodal ensemble (nnU-Net+DINOv3+Clinical) achieved the highest external AUC (0.742), exceeding radiomics (ΔAUC +0.114, p = 0.008) but not nnU-Net (ΔAUC +0.011, p = 0.737). Compared with the clinical model, discrimination was higher but nonsignificant at the prespecified threshold (ΔAUC +0.098, p = 0.050), while reclassification improved (IDI +0.109, p < 0.001). Radiomics-containing combinations performed strongly in development but lost most discrimination externally. The multimodal ensemble had an external calibration slope of 1.32 (Brier 0.190) and net benefit above default strategies at thresholds ≥ 0.23. The multimodal ensemble achieved the highest external AUC but did not significantly exceed nnU-Net, which was the most transportable single predictor of meningioma grade. Clinical variables did not improve discrimination but enabled interpretable feature-level attribution. Handcrafted radiomics lacked external transportability.