Enhancing Medical Image Segmentation via Heat Conduction Equation

2025-11-05

AI Summary

In medical image segmentation, existing U-Net–based models struggle to simultaneously capture global contextual information and long-range dependencies under limited computational budgets. To address this, we propose U-Mamba—a novel hybrid architecture integrating the Mamba state-space module with a newly introduced Heat Conduction Operator (HCO). HCO, formulated in the frequency domain, explicitly models thermal diffusion dynamics, offering an interpretable mechanism for global feature propagation; Mamba efficiently captures long-range spatial dependencies. Their synergistic integration enhances semantic representation while significantly improving computational efficiency. Evaluated on multi-modal abdominal CT/MRI datasets, U-Mamba consistently outperforms state-of-the-art baselines across multiple metrics, demonstrating superior accuracy, generalizability, and practical applicability.

Technology Category

• Computer Vision: Segmentation
• Machine Learning: Calibration & Uncertainty Quantification
• Search and Optimization: Learning to Search

Application Category

• 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
• Graph Algorithms and Modeling for the Web: Efficient manipulation of static and dynamic Web-related graphs

Abstract

Medical image segmentation has been significantly advanced by deep learning architectures, notably U-Net variants. However, existing models struggle to achieve efficient global context modeling and long-range dependency reasoning under practical computational budgets simultaneously. In this work, we propose a novel hybrid architecture utilizing U-Mamba with Heat Conduction Equation. Our model combines Mamba-based state-space modules for efficient long-range reasoning with Heat Conduction Operators (HCOs) in the bottleneck layers, simulating frequency-domain thermal diffusion for enhanced semantic abstraction. Experimental results on multimodal abdominal CT and MRI datasets demonstrate that the proposed model consistently outperforms strong baselines, validating its effectiveness and generalizability. It suggest that blending state-space dynamics with heat-based global diffusion offers a scalable and interpretable solution for medical segmentation tasks.

Problem

• Improving global context modeling in medical image segmentation
• Addressing long-range dependency reasoning under computational constraints
• Enhancing semantic abstraction through heat conduction simulation

Innovation

• U-Mamba with Heat Conduction Equation hybrid architecture
• Mamba state-space modules enable efficient long-range reasoning
• Heat Conduction Operators simulate thermal diffusion for abstraction