CAPE: Connectivity-Aware Path Enforcement Loss for Curvilinear Structure Delineation
This addresses the challenge of preserving topological correctness in biomedical imaging for applications such as neuron and blood vessel tracing, representing an incremental advance in loss function design.
The paper tackles the problem of poor connectivity in semantic segmentation of curvilinear structures like neurons and blood vessels by proposing CAPE, a novel loss function that enforces connectivity in graphs derived from segmentation maps, resulting in significant improvements in topology-aware metrics and outperforming state-of-the-art methods.
Promoting the connectivity of curvilinear structures, such as neuronal processes in biomedical scans and blood vessels in CT images, remains a key challenge in semantic segmentation. Traditional pixel-wise loss functions, including cross-entropy and Dice losses, often fail to capture high-level topological connectivity, resulting in topological mistakes in graphs obtained from prediction maps. In this paper, we propose CAPE (Connectivity-Aware Path Enforcement), a novel loss function designed to enforce connectivity in graphs obtained from segmentation maps by optimizing a graph connectivity metric. CAPE uses the graph representation of the ground truth to select node pairs and determine their corresponding paths within the predicted segmentation through a shortest-path algorithm. Using this, we penalize both disconnections and false positive connections, effectively promoting the model to preserve topological correctness. Experiments on 2D and 3D datasets, including neuron and blood vessel tracing demonstrate that CAPE significantly improves topology-aware metrics and outperforms state-of-the-art methods.