DSVM-UNet : Enhancing VM-UNet with Dual Self-distillation for Medical Image Segmentation

2026-01-27

AI Summary

This work addresses the challenge of balancing model performance and computational efficiency in medical image segmentation by proposing a structure-preserving dual self-distillation mechanism built upon the VM-UNet architecture. The method enhances semantic awareness through alignment of global and local feature representations, leveraging Vision Mamba’s strength in efficiently modeling long-range dependencies. Without altering the underlying network structure, the approach achieves state-of-the-art performance on the ISIC2017, ISIC2018, and Synapse datasets while maintaining high computational efficiency, thereby effectively reconciling accuracy and resource consumption.

Technology Category

• Computer Vision: Segmentation
• Machine Learning: Deep Neural Architectures and Foundation Models
• 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
• Graph Algorithms and Modeling for the Web: Foundation models and LLMs for Web-related graphs
• Search and Retrieval-Augmented AI: Efficiency and scalability of Web search engines

Abstract

Vision Mamba models have been extensively researched in various fields, which address the limitations of previous models by effectively managing long-range dependencies with a linear-time overhead. Several prospective studies have further designed Vision Mamba based on UNet(VM-UNet) for medical image segmentation. These approaches primarily focus on optimizing architectural designs by creating more complex structures to enhance the model's ability to perceive semantic features. In this paper, we propose a simple yet effective approach to improve the model by Dual Self-distillation for VM-UNet (DSVM-UNet) without any complex architectural designs. To achieve this goal, we develop double self-distillation methods to align the features at both the global and local levels. Extensive experiments conducted on the ISIC2017, ISIC2018, and Synapse benchmarks demonstrate that our approach achieves state-of-the-art performance while maintaining computational efficiency. Code is available at https://github.com/RoryShao/DSVM-UNet.git.

Problem

• Medical Image Segmentation
• Vision Mamba
• VM-UNet
• Model Performance
• Architectural Complexity

Innovation

• Dual Self-distillation
• VM-UNet
• Medical Image Segmentation
• Vision Mamba
• Feature Alignment