Shape-aware Sampling Matters in the Modeling of Multi-Class Tubular Structures

2025-06-14

AI Summary

Existing methods model multi-class tubular structures using voxel-overlap accuracy, neglecting their fine-grained geometric and topological complexity—leading to topological distortions such as branch discontinuities and poor connectivity. To address this, we propose a shape-aware, topology-preserving modeling framework. Our method introduces Fractal Dimension-driven Patch Sizing (FDPS), the first adaptive patch-size allocation strategy guided by fractal dimension to enable structure-sensitive sampling. It further integrates Minimum Path Cost Skeletonization (MPC-Skel) with a skeleton-weighted loss to explicitly enforce topological consistency. Evaluated on two tubular datasets, our approach achieves significant improvements: Dice score increases substantially, branch-point error decreases by 32.7%, and connectivity score improves by 28.4%. These results demonstrate superior morphological fidelity and topological integrity over prior methods.

Technology Category

• Machine Learning: Learning with Manifolds
• Computer Vision: Segmentation
• Search and Optimization: Mixed Discrete/Continuous Search

Application Category

• Graph Algorithms and Modeling for the Web: Efficient manipulation of static and dynamic Web-related graphs
• Web Mining and Content Analysis: Robustness and generalizability of Web mining methods
• User Modeling, Personalization and Recommendation: Federated recommendation systems and personalization

Abstract

Accurate multi-class tubular modeling is critical for precise lesion localization and optimal treatment planning. Deep learning methods enable automated shape modeling by prioritizing volumetric overlap accuracy. However, the inherent complexity of fine-grained semantic tubular shapes is not fully emphasized by overlap accuracy, resulting in reduced topological preservation. To address this, we propose the Shapeaware Sampling (SAS), which optimizes patchsize allocation for online sampling and extracts a topology-preserved skeletal representation for the objective function. Fractal Dimension-based Patchsize (FDPS) is first introduced to quantify semantic tubular shape complexity through axis-specific fractal dimension analysis. Axes with higher fractal complexity are then sampled with smaller patchsizes to capture fine-grained features and resolve structural intricacies. In addition, Minimum Path-Cost Skeletonization (MPC-Skel) is employed to sample topologically consistent skeletal representations of semantic tubular shapes for skeleton-weighted objective functions. MPC-Skel reduces artifacts from conventional skeletonization methods and directs the focus to critical topological regions, enhancing tubular topology preservation. SAS is computationally efficient and easily integrable into optimization pipelines. Evaluation on two semantic tubular datasets showed consistent improvements in both volumetric overlap and topological integrity metrics.

Problem

• Improving multi-class tubular shape modeling accuracy
• Addressing topological preservation in semantic tubular structures
• Optimizing patch-size allocation for fine-grained feature capture

Innovation

• Shapeaware Sampling optimizes patchsize allocation
• Fractal Dimension quantifies tubular shape complexity
• Minimum Path-Cost Skeletonization preserves topology