Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

HC-Mamba: Vision MAMBA with Hybrid Convolutional Techniques for Medical Image Segmentation

Jiashu Xu

TL;DR

The paper tackles downsampling-induced information loss in medical image segmentation by introducing HC-Mamba, a hybrid convolution model built on a state-space model backbone. It combines dilated convolutions to widen the receptive field with depthwise separable convolutions to cut parameters, organized within the HC-SSM framework and featuring the SS2D module and a two-branch HC-SSM Block. A weighted loss combining mIoU, Dice, and boundary terms guides training to optimize overlap, similarity, and boundary precision. Experiments on Synapse, ISIC2017, and ISIC2018 demonstrate competitive performance and substantial parameter efficiency (roughly a 50–60% reduction compared with some Mamba-based models), with ablations confirming the contributions of both dilated and depthwise operations. The work offers a practical, scalable approach for accurate medical image segmentation suitable for real-time and large-scale clinical data analysis.

Abstract

Automatic medical image segmentation technology has the potential to expedite pathological diagnoses, thereby enhancing the efficiency of patient care. However, medical images often have complex textures and structures, and the models often face the problem of reduced image resolution and information loss due to downsampling. To address this issue, we propose HC-Mamba, a new medical image segmentation model based on the modern state space model Mamba. Specifically, we introduce the technique of dilated convolution in the HC-Mamba model to capture a more extensive range of contextual information without increasing the computational cost by extending the perceptual field of the convolution kernel. In addition, the HC-Mamba model employs depthwise separable convolutions, significantly reducing the number of parameters and the computational power of the model. By combining dilated convolution and depthwise separable convolutions, HC-Mamba is able to process large-scale medical image data at a much lower computational cost while maintaining a high level of performance. We conduct comprehensive experiments on segmentation tasks including organ segmentation and skin lesion, and conduct extensive experiments on Synapse, ISIC17 and ISIC18 to demonstrate the potential of the HC-Mamba model in medical image segmentation. The experimental results show that HC-Mamba exhibits competitive performance on all these datasets, thereby proving its effectiveness and usefulness in medical image segmentation.

HC-Mamba: Vision MAMBA with Hybrid Convolutional Techniques for Medical Image Segmentation

TL;DR

The paper tackles downsampling-induced information loss in medical image segmentation by introducing HC-Mamba, a hybrid convolution model built on a state-space model backbone. It combines dilated convolutions to widen the receptive field with depthwise separable convolutions to cut parameters, organized within the HC-SSM framework and featuring the SS2D module and a two-branch HC-SSM Block. A weighted loss combining mIoU, Dice, and boundary terms guides training to optimize overlap, similarity, and boundary precision. Experiments on Synapse, ISIC2017, and ISIC2018 demonstrate competitive performance and substantial parameter efficiency (roughly a 50–60% reduction compared with some Mamba-based models), with ablations confirming the contributions of both dilated and depthwise operations. The work offers a practical, scalable approach for accurate medical image segmentation suitable for real-time and large-scale clinical data analysis.

Abstract

Automatic medical image segmentation technology has the potential to expedite pathological diagnoses, thereby enhancing the efficiency of patient care. However, medical images often have complex textures and structures, and the models often face the problem of reduced image resolution and information loss due to downsampling. To address this issue, we propose HC-Mamba, a new medical image segmentation model based on the modern state space model Mamba. Specifically, we introduce the technique of dilated convolution in the HC-Mamba model to capture a more extensive range of contextual information without increasing the computational cost by extending the perceptual field of the convolution kernel. In addition, the HC-Mamba model employs depthwise separable convolutions, significantly reducing the number of parameters and the computational power of the model. By combining dilated convolution and depthwise separable convolutions, HC-Mamba is able to process large-scale medical image data at a much lower computational cost while maintaining a high level of performance. We conduct comprehensive experiments on segmentation tasks including organ segmentation and skin lesion, and conduct extensive experiments on Synapse, ISIC17 and ISIC18 to demonstrate the potential of the HC-Mamba model in medical image segmentation. The experimental results show that HC-Mamba exhibits competitive performance on all these datasets, thereby proving its effectiveness and usefulness in medical image segmentation.
Paper Structure (13 sections, 6 equations, 4 figures, 3 tables)

This paper contains 13 sections, 6 equations, 4 figures, 3 tables.

Table of Contents

  1. Introduction
  2. Methods
  3. Preliminaries
  4. Model structure
  5. SS2D module
  6. HC-SSM Block
  7. Loss function
  8. Experiments
  9. Datasets
  10. Results
  11. Ablation experiments
  12. Discussion
  13. Conclusion

Figures (4)

  • Figure 1: SSM(state space model)process diagram
  • Figure 2: (a) Overall structure of HC-Mamba. (b) Overall structure of HC-SSM Bloc
  • Figure 3: Comparison diagram between expansion rate of 1,2,3 (left) and expansion rate of 2,2,2 (right)
  • Figure 4: Receptive field diagram using three layers of ordinary convolution