BiM-GeoAttn-Net: Linear-Time Depth Modeling with Geometry-Aware Attention for 3D Aortic Dissection CTA Segmentation

2026-02-27

AI Summary

This study addresses the challenges of inter-slice inconsistency and ambiguous boundaries in 3D aortic dissection segmentation from CTA images, which stem from insufficient long-range contextual modeling and low image contrast. To overcome these issues, the authors propose BiM-GeoAttn-Net, a lightweight framework that uniquely integrates a linear-complexity state space model with a direction-sensitive anisotropic geometric attention mechanism. Specifically, bidirectional deep Mamba modules efficiently capture cross-slice dependencies, while a geometry-aware vascular attention module refines tubular structures and sharpens boundary delineation. Evaluated on multi-source aortic dissection CTA datasets, the method achieves a Dice score of 93.35% and an HD95 of 12.36 mm, significantly outperforming current CNN-, Transformer-, and state space model–based approaches in both volumetric overlap and boundary accuracy.

Technology Category

• Computer Vision: Segmentation
• Knowledge Representation and Reasoning: Geometric, Spatial, and Temporal Reasoning
• Constraint Satisfaction and Optimization: Satisfiability Modulo Theories

Application Category

• Graph Algorithms and Modeling for the Web: Algorithms and analysis for heterogeneous, signed, attributed, multi-relational, temporal, higher-order, and annotated Web-related graphs
• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs
• Search and Retrieval-Augmented AI: Retrieval-Augmented Generation (RAG) and multi-modal RAG

Abstract

Accurate segmentation of aortic dissection (AD) lumens in CT angiography (CTA) is essential for quantitative morphological assessment and clinical decision-making. However, reliable 3D delineation remains challenging due to limited long-range context modeling, which compromises inter-slice coherence, and insufficient structural discrimination under low-contrast conditions. To address these limitations, we propose BiM-GeoAttn-Net, a lightweight framework that integrates linear-time depth-wise state-space modeling with geometry-aware vessel refinement. Our approach is featured by Bidirectional Depth Mamba (BiM) to efficiently capture cross-slice dependencies and Geometry-Aware Vessel Attention (GeoAttn) module that employs orientation-sensitive anisotropic filtering to refine tubular structures and sharpen ambiguous boundaries. Extensive experiments on a multi-source AD CTA dataset demonstrate that BiM-GeoAttn-Net achieves a Dice score of 93.35% and an HD95 of 12.36 mm, outperforming representative CNN-, Transformer-, and SSM-based baselines in overlap metrics while maintaining competitive boundary accuracy. These results suggest that coupling linear-time depth modeling with geometry-aware refinement provides an effective, computationally efficient solution for robust 3D AD segmentation.

Problem

• aortic dissection
• 3D segmentation
• CT angiography
• long-range context
• low-contrast

Innovation

• Bidirectional Depth Mamba
• Geometry-Aware Attention
• Linear-Time Depth Modeling
• Aortic Dissection Segmentation
• Anisotropic Filtering