Deep learning improves brain surface labeling with limited expert data
Geometric-to-Semantic Spherical Transfer Learning for Cortical Sulci Labeling
Machine LearningComputer Vision and Pattern Recognition
Summary
Labeling the grooves on the brain’s surface is hard because there are many small, complex features and only a few expert examples to learn from. The authors used a huge set of unlabeled brain scans to train a model to understand brain shape before teaching it to recognize specific labeled grooves. This two-step approach helps the model identify rare and tricky brain features better than training with limited labeled data alone. Their method increased accuracy notably, especially on the most variable and uncommon brain grooves.
What this means in practice
- •For medical image analysts: Improve automated labeling of brain sulci in MRI scans where expert annotations are scarce by using pretrained geometric features.
- •For neuroimaging software developers: Enhance cortical surface analysis tools by integrating pretrained geometric priors to boost identification of rare and small brain features.
Authors
Saeb Tounsi, Joël Chavas, Pietro Gori, Vincent Frouin, Denis Rivière, Jean-François Mangin
Abstract
Deep learning on cortical surfaces faces a dilemma: capturing the complex topology of over 60 nomenclature-dependent sulci per hemisphere requires high-capacity models, yet the extreme scarcity of expert annotations ($N=62$ subjects) inevitably causes overfitting. Standard supervised approaches fail to generalize in this data-scarce regime, particularly for variable and small sulci where topological ambiguity is high. To overcome this limitation, we introduce a Geometric-to-Semantic Spherical Transfer Learning framework. First, we leverage massive unlabeled data (UK Biobank, $\approx$30,000 subjects) to pre-train a spherical encoder using a locally-optimized strategy. By relying solely on continuous surface features (curvature and depth), the relevance of this pre-training is confirmed by the model's ability to detect localized and rare topological traits, such as sulcal interruptions. The downstream labeling task, however, introduces extracted sulcal fundi (lines) as an explicit semantic input. To bridge this dimensional domain gap (from purely geometric to semantic) without causing catastrophic forgetting, these anatomical lines are integrated into the pre-trained backbone via a soft-initialized Topological Prior Injector. Our experiments demonstrate that this approach outperforms fully supervised baselines trained from scratch, achieving a mean Dice of 0.77. Crucially, a local analysis reveals that the self-supervised geometric priors yield the largest performance gains on variable and tertiary sulci (up to 14.8%), confirming that learning the cortex shape is highly beneficial for identifying its rarest parts.