Noninvasive molecular subtyping of pediatric low-grade glioma with self-supervised transfer learning

Abstract

ABSTRACTKey ResultsAn innovative training approach combining self-supervision and transfer learning (“TransferX”) is developed to boost model performance in a low data setting;TransferX enables the development of a scan-to-prediction pipeline for pediatric LGG mutational status (BRAF V600E, fusion, or wildtype) with ≥75% accuracy on internal and external validation;An evaluation metric, “COMDist”, is introduced to increase interpretability and quantify the accuracy of the model’s attention around the tumor.PurposeTo develop and externally validate a scan-to-prediction deep-learning pipeline for noninvasive, MRI-based BRAF mutational status classification for pLGG.Materials and MethodsWe conducted a retrospective study of two pLGG datasets with linked genomic and diagnostic T2-weighted MRI of patients: BCH (development dataset, n=214 [60 (28%) BRAF fusion, 50 (23%) BRAF V600E, 104 (49%) wild-type), and Child Brain Tumor Network (CBTN) (external validation, n=112 [60 (53%) BRAF-Fusion, 17 (15%) BRAF-V600E, 35 (32%) wild-type]). We developed a deep learning pipeline to classify BRAF mutational status (V600E vs. fusion vs. wild-type) via a two-stage process: 1) 3D tumor segmentation and extraction of axial tumor images, and 2) slice-wise, deep learning-based classification of mutational status. We investigated knowledge-transfer and self-supervised approaches to prevent model overfitting with a primary endpoint of the area under the receiver operating characteristic curve (AUC). To enhance model interpretability, we developed a novel metric, COMDist, that quantifies the accuracy of model attention around the tumor.ResultsA combination of transfer learning from a pretrained medical imaging-specific network and self-supervised label cross-training (TransferX) coupled with consensus logic yielded the highest macro-average AUC (0.82 [95% CI: 0.70-0.90]) and accuracy (77%) on internal validation, with an AUC improvement of +17.7% and a COMDist improvement of +6.4% versus training from scratch. On external validation, the TransferX model yielded AUC (0.73 [95% CI 0.68-0.88]) and accuracy (75%).ConclusionTransfer learning and self-supervised cross-training improved classification performance and generalizability for noninvasive pLGG mutational status prediction in a limited data scenario.