Addressing data annotation scarcity in Brain Tumor Segmentation on 3D MRI scan Using a Semi-Supervised Teacher-Student Framework
This addresses the challenge of expensive data annotation and heterogeneity in medical imaging for clinicians and researchers, with incremental improvements in semi-supervised learning methods.
The paper tackled the problem of brain tumor segmentation from 3D MRI scans with scarce annotations by proposing a semi-supervised teacher-student framework, resulting in a validation Dice score increase from 0.393 with 10% data to 0.872 with 100% data and the student surpassing the teacher on specific tumor subregions.
Accurate brain tumor segmentation from MRI is limited by expensive annotations and data heterogeneity across scanners and sites. We propose a semi-supervised teacher-student framework that combines an uncertainty-aware pseudo-labeling teacher with a progressive, confidence-based curriculum for the student. The teacher produces probabilistic masks and per-pixel uncertainty; unlabeled scans are ranked by image-level confidence and introduced in stages, while a dual-loss objective trains the student to learn from high-confidence regions and unlearn low-confidence ones. Agreement-based refinement further improves pseudo-label quality. On BraTS 2021, validation DSC increased from 0.393 (10% data) to 0.872 (100%), with the largest gains in early stages, demonstrating data efficiency. The teacher reached a validation DSC of 0.922, and the student surpassed the teacher on tumor subregions (e.g., NCR/NET 0.797 and Edema 0.980); notably, the student recovered the Enhancing class (DSC 0.620) where the teacher failed. These results show that confidence-driven curricula and selective unlearning provide robust segmentation under limited supervision and noisy pseudo-labels.