Explainable Continuous-Time Mask Refinement with Local Self-Similarity Priors for Medical Image Segmentation

2026-02-27

AI Summary

This work addresses the challenge of boundary segmentation in medical images of foot ulcers, where tissue heterogeneity and low contrast impede accurate delineation. The authors propose LSS-LTCNet, a novel framework that uniquely integrates local self-similarity (LSS) texture priors with a liquid time-constant (LTC) continuous-time dynamical system governed by ordinary differential equations (ODEs). This ante-hoc interpretable architecture explicitly differentiates necrotic tissue from background and refines boundary evolution through an ODE-driven iterative mechanism, offering an intrinsic visual audit trail. Evaluated on the MICCAI FUSeg dataset, the model achieves a Dice score of 86.96% and a Hausdorff distance (HD95) of 8.91 pixels with only 25.70 million parameters, outperforming both U-Net and Transformer-based baselines.

Technology Category

• Computer Vision: Segmentation
• Machine Learning: Large Multimodal Models (LMMs)
• Search and Optimization: Learning to Search

Application Category

• Graph Algorithms and Modeling for the Web: Foundation models and LLMs for Web-related graphs
• Systems and Infrastructure for Web, Mobile and WoT: Experiences and lessons learnt from Web-based algorithms and system deployments
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking

Abstract

Accurate semantic segmentation of foot ulcers is essential for automated wound monitoring, yet boundary delineation remains challenging due to tissue heterogeneity and poor contrast with surrounding skin. To overcome the limitations of standard intensity-based networks, we present LSS-LTCNet:an ante-hoc explainable framework synergizing deterministic structural priors with continuous-time neural dynamics. Our architecture departs from traditional black-box models by employing a Local Self-Similarity (LSS) mechanism that extracts dense, illumination-invariant texture descriptors to explicitly disentangle necrotic tissue from background artifacts. To enforce topological precision, we introduce a Liquid Time-Constant (LTC) refinement module that treats boundary evolution as an ODEgoverned dynamic system, iteratively refining masks over continuous time-steps. Comprehensive evaluation on the MICCAI FUSeg dataset demonstrates that LSS-LTCNet achieves state-of-the-art boundary alignment, securing a peak Dice score of 86.96% and an exceptional 95th percentile Hausdorff Distance (HD95) of 8.91 pixels. Requiring merely 25.70M parameters, the model significantly outperforms heavier U-Net and transformer baselines in efficiency. By providing inherent visual audit trails alongside high-fidelity predictions, LSS-LTCNet offers a robust and transparent solution for computer-aided diagnosis in mobile healthcare (mHealth) settings.

Problem

• medical image segmentation
• foot ulcer
• boundary delineation
• tissue heterogeneity
• low contrast

Innovation

• Local Self-Similarity
• Liquid Time-Constant
• Continuous-Time Neural Dynamics
• Explainable AI
• Medical Image Segmentation