Anatomically Conditioned Recurrent Refinement for Topology-Aware Circle of Willis Segmentation
For medical image segmentation of complex vascular structures, this method improves topological connectivity without sacrificing volumetric accuracy.
The paper tackles the problem of fragmented Circle of Willis segmentation from MRA by proposing AC2RUNet, which reduces Hausdorff Distance from 9.17 mm to 4.72 mm and Betti number errors from 0.40 to 0.19 compared to nnU-Net.
Segmenting the Circle of Willis (CoW) from Magnetic Resonance Angiography (MRA) is challenging due to complex topology and thin vascular structures that are prone to fragmentation. Standard Convolutional Neural Networks (CNNs) often fail to capture these topological constraints, resulting in "broken vessel" artifacts. To address this, we propose the Anatomically Conditioned Recurrent Refinement U-Net (AC2RUNet). Our architecture decouples segmentation into two streams: a Static Stream that extracts invariant anatomical features and a lightweight Dynamic Stream that iteratively refines topological errors over time. We further introduce a dynamic curriculum learning strategy that transitions from high-recall geometric supervision to topology-aware constraints. Validated on the TopCoW dataset, AC2RUNet substantially reduces Hausdorff Distance (4.72 mm vs 9.17 mm) and Betti number errors (0.19 vs 0.40), improving topological connectivity over the nnU-Net baseline while maintaining comparable volumetric Dice.