SkinMamba: A Precision Skin Lesion Segmentation Architecture with Cross-Scale Global State Modeling and Frequency Boundary Guidance

TL;DR

SkinMamba is a novel hybrid architecture for skin lesion segmentation that combines efficient global and local feature modeling with boundary guidance, achieving high accuracy with linear complexity.

Contribution

The paper introduces SkinMamba, a hybrid CNN-Transformer model with a novel SRSSB and FBGM modules, enabling effective global context modeling and boundary precision in skin lesion segmentation.

Findings

Outperforms existing methods on ISIC datasets

Maintains linear complexity while modeling global dependencies

Achieves accurate boundary segmentation

Abstract

Skin lesion segmentation is a crucial method for identifying early skin cancer. In recent years, both convolutional neural network (CNN) and Transformer-based methods have been widely applied. Moreover, combining CNN and Transformer effectively integrates global and local relationships, but remains limited by the quadratic complexity of Transformer. To address this, we propose a hybrid architecture based on Mamba and CNN, called SkinMamba. It maintains linear complexity while offering powerful long-range dependency modeling and local feature extraction capabilities. Specifically, we introduce the Scale Residual State Space Block (SRSSB), which captures global contextual relationships and cross-scale information exchange at a macro level, enabling expert communication in a global state. This effectively addresses challenges in skin lesion segmentation related to varying lesion sizes and inconspicuous target areas. Additionally, to mitigate boundary blurring and information loss during model downsampling, we introduce the Frequency Boundary Guided Module (FBGM), providing sufficient boundary priors to guide precise boundary segmentation, while also using the retained information to assist the decoder in the decoding process. Finally, we conducted comparative and ablation experiments on two public lesion segmentation datasets (ISIC2017 and ISIC2018), and the results demonstrate the strong competitiveness of SkinMamba in skin lesion segmentation tasks. The code is available at https://github.com/zs1314/SkinMamba.