AI and Cancer Research ── Brain ── 2026-11-12
Brain tumours: treatment, trials and AI
- Abstract
Public brain metastasis MRI cohorts are widely used for prognostic modeling, but whether an endpoint is identifiable at the available sample size is rarely tested. We evaluated eight clinical endpoints in two public cohorts, Yale/BraTS-MET (199 patients, 74 deaths) and OpenBTAI/MOLAB (75 patients, all deceased), using lesion-centered nnU-Net encoder embeddings and PyRadiomics features with patient-grouped cross-validation and label permutation. Patient-level infratentorial involvement served as a positive control (balanced accuracy 0.60); the Graded Prognostic Assessment (GPA) score was recoverable from OpenBTAI radiomics (ρ = 0.42), whereas brain age was not recovered in either cohort. Primary tumor type was the only endpoint significant after multiplicity correction (1.3× chance). Survival endpoints were null for clinical models and the planned graph model (C = 0.50, p = 0.49); an exploratory linear model on embeddings reached C ≈ 0.58 (p = 0.07, uncorrected), which PyRadiomics features did not reproduce. No progression definition remained significant after family-wise correction. In simulation, 80% power required a true C-index of about 0.68–0.76, so null survival results cannot exclude moderate effects. GPA did not stratify survival (p = 0.18) in a cohort containing only deceased patients. An extraction error found during revision was corrected.
Journal IF-equivalent: 4.3 (OpenAlex 2-year mean citedness, value as of 2026-10-15, retrieved 2026-10-16; not the official Clarivate IF)Reference: Bao X, Wang J, Zhao X. Survival Prediction from Public Brain Metastasis MRI Cohorts: A Positive-Control and Permutation-Based Feasibility Evaluation. J Imaging. 2026 Nov 8;12(11):560. doi:10.3390/jimaging12110560.Checked: Abstract only - Abstract
Postoperative magnetic resonance imaging (MRI) captures anatomical and tissue characteristics relevant to glioma prognosis. We evaluated the contribution of compartment-specific imaging features to clinical and molecular prediction using MU-Glioma-Post. Of 187 patients with usable radiomics, 111 were eligible for 180-day recurrence/progression prediction and 114 were eligible for 365-day mortality prediction after the first available postoperative MRI (TP1). Clinical-only, radiomics-only, and combined elastic-net logistic models were assessed using nested cross-validation and a locked internal holdout. For recurrence, the combined model achieved development out-of-fold (OOF) and holdout areas under the receiver operating characteristic curve (AUROCs) of 0.742 and 0.786 (95% confidence interval, 0.562–0.959), respectively. Morphology was the highest-ranking imaging-only configuration (OOF AUROC, 0.724), and flatness of the surrounding non-enhancing FLAIR hyperintensity led the SHAP feature ranking. In exploratory repeated nested cross-validation, adding shape, volume, and region-presence features to clinical predictors increased mean recurrence AUROC from 0.743 to 0.794, compared with 0.747 for clinical predictors combined with full radiomics. Clinical predictors remained the main source of mortality discrimination, with mean AUROCs of 0.774 for clinical predictors, 0.777 with morphology, and 0.768 with full radiomics. These findings identify postoperative geometry as a promising source of recurrence-related information and support focused integration of morphological features with clinical predictors.
Journal IF-equivalent: 7.1 (OpenAlex 2-year mean citedness, value as of 2026-11-09, retrieved 2026-11-11; not the official Clarivate IF)Reference: Farokhi Nia S, Shafaie V, Movahedi Rad M. Clinical-Radiomic Integration for Postoperative Glioma Recurrence and Survival Prediction. Mach Learn Knowl Extr. 2026 Nov 8;8(11):372. doi:10.3390/make8110372.Checked: Abstract only