Topograph: An efficient Graph-Based Framework for Strictly Topology Preserving Image Segmentation
This addresses the need for topologically correct segmentation in medical imaging and other domains, offering a computationally efficient solution with formal guarantees.
The paper tackles the problem of topological inaccuracy in image segmentation by proposing a graph-based framework that ensures strict topology preservation. The method achieves state-of-the-art performance with up to fivefold faster loss computation compared to existing persistent homology approaches.
Topological correctness plays a critical role in many image segmentation tasks, yet most networks are trained using pixel-wise loss functions, such as Dice, neglecting topological accuracy. Existing topology-aware methods often lack robust topological guarantees, are limited to specific use cases, or impose high computational costs. In this work, we propose a novel, graph-based framework for topologically accurate image segmentation that is both computationally efficient and generally applicable. Our method constructs a component graph that fully encodes the topological information of both the prediction and ground truth, allowing us to efficiently identify topologically critical regions and aggregate a loss based on local neighborhood information. Furthermore, we introduce a strict topological metric capturing the homotopy equivalence between the union and intersection of prediction-label pairs. We formally prove the topological guarantees of our approach and empirically validate its effectiveness on binary and multi-class datasets. Our loss demonstrates state-of-the-art performance with up to fivefold faster loss computation compared to persistent homology methods.