VasGuideNet: Vascular Topology-Guided Couinaud Liver Segmentation with Structural Contrastive Loss

2026-02-24

AI Summary

Existing Couinaud liver segmentation methods often yield ambiguous boundaries and exhibit limited generalization under anatomical variations due to their neglect of vascular topology. To address this, this work proposes VasGuideNet, the first framework to explicitly integrate vascular topology into the segmentation process. It employs a graph convolutional network to model the geometric and connectivity characteristics of the vascular skeleton and fuses this information into a 3D encoder–decoder architecture via a cross-attention mechanism. Additionally, a structural contrastive loss and a global memory bank are introduced to enhance anatomical consistency and inter-class separability. Evaluated on the Task08_HepaticVessel and LASSD datasets, VasGuideNet achieves Dice scores of 83.68% and 76.65%, with relative volume differences (RVD) of 1.68 and 7.08, respectively, outperforming baseline models such as UNETR and Swin UNETR.

Technology Category

• Computer Vision: Segmentation
• Machine Learning: Deep Generative Models & Autoencoders
• Reasoning under Uncertainty: Graphical Models

Application Category

• Graph Algorithms and Modeling for the Web: Graph neural networks and deep learning approaches for Web-related graphs
• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs
• Web Mining and Content Analysis: Large pretrained models with web data

Abstract

Accurate Couinaud liver segmentation is critical for preoperative surgical planning and tumor localization.However, existing methods primarily rely on image intensity and spatial location cues, without explicitly modeling vascular topology. As a result, they often produce indistinct boundaries near vessels and show limited generalization under anatomical variability.We propose VasGuideNet, the first Couinaud segmentation framework explicitly guided by vascular topology. Specifically, skeletonized vessels, Euclidean distance transform (EDT)--derived geometry, and k-nearest neighbor (kNN) connectivity are encoded into topology features using Graph Convolutional Networks (GCNs). These features are then injected into a 3D encoder--decoder backbone via a cross-attention fusion module. To further improve inter-class separability and anatomical consistency, we introduce a Structural Contrastive Loss (SCL) with a global memory bank.On Task08_HepaticVessel and our private LASSD dataset, VasGuideNet achieves Dice scores of 83.68% and 76.65% with RVDs of 1.68 and 7.08, respectively. It consistently outperforms representative baselines including UNETR, Swin UNETR, and G-UNETR++, delivering higher Dice/mIoU and lower RVD across datasets, demonstrating its effectiveness for anatomically consistent segmentation. Code is available at https://github.com/Qacket/VasGuideNet.git.

Problem

• Couinaud segmentation
• vascular topology
• anatomical variability
• boundary ambiguity
• liver segmentation

Innovation

• vascular topology
• Graph Convolutional Networks
• Structural Contrastive Loss
• Couinaud segmentation
• cross-attention fusion