Centerline Boundary Dice Loss for Vascular Segmentation

TL;DR

This paper introduces the cbDice loss function for vascular segmentation, combining topological preservation with geometric detail recognition, and demonstrates its superior performance across multiple datasets including challenging 3D and multi-class cases.

Contribution

The paper proposes the cbDice loss, which integrates boundary awareness and vessel radius information to improve vascular segmentation accuracy and robustness.

Findings

cbDice outperforms traditional clDice and B-DoU losses.

It maintains topological and geometric consistency across vessel sizes.

Validated on diverse datasets, including MICCAI 2023 TopCoW Challenge.

Abstract

Vascular segmentation in medical imaging plays a crucial role in analysing morphological and functional assessments. Traditional methods, like the centerline Dice (clDice) loss, ensure topology preservation but falter in capturing geometric details, especially under translation and deformation. The combination of clDice with traditional Dice loss can lead to diameter imbalance, favoring larger vessels. Addressing these challenges, we introduce the centerline boundary Dice (cbDice) loss function, which harmonizes topological integrity and geometric nuances, ensuring consistent segmentation across various vessel sizes. cbDice enriches the clDice approach by including boundary-aware aspects, thereby improving geometric detail recognition. It matches the performance of the boundary difference over union (B-DoU) loss through a mask-distance-based approach, enhancing traslation sensitivity. Crucially, cbDice incorporates radius information from vascular skeletons, enabling uniform adaptation to vascular diameter changes and maintaining balance in branch growth and fracture impacts. Furthermore, we conducted a theoretical analysis of clDice variants (cl-X-Dice). We validated cbDice's efficacy on three diverse vascular segmentation datasets, encompassing both 2D and 3D, and binary and multi-class segmentation. Particularly, the method integrated with cbDice demonstrated outstanding performance on the MICCAI 2023 TopCoW Challenge dataset. Our code is made publicly available at: https://github.com/PengchengShi1220/cbDice.