MambaLiteUNet: Cross-Gated Adaptive Feature Fusion for Robust Skin Lesion Segmentation

2026-04-22

AI Summary

This study addresses the challenge that existing lightweight skin lesion segmentation models struggle to accurately delineate lesion boundaries and textures, thereby hindering early skin cancer diagnosis. To overcome this limitation, the authors propose a lightweight U-Net architecture grounded in the Mamba state space model, incorporating three novel components: Adaptive Multi-branch Fusion (AMF), Local-Global Feature Mixing (LGFM), and Cross-Gated Attention (CGA). These modules enhance feature interaction while preserving fine-grained spatial details. Despite drastically reducing model size—achieving a 93.6% parameter reduction and 97.6% lower computational cost—the method attains an average IoU of 87.12% and Dice score of 93.09% across multiple datasets, while also demonstrating strong out-of-domain generalization capabilities.

Technology Category

• Computer Vision: Segmentation
• Machine Learning: Learning on the Edge & Model Compression
• Search and Optimization: Learning to Search

Application Category

• Web Mining and Content Analysis: Large pretrained models with web data
• Systems and Infrastructure for Web, Mobile and WoT: Applied ML and AI for Web-based mobile applications
• User Modeling, Personalization and Recommendation: On-Device user modeling, personalization, and recommendation

Abstract

Recent segmentation models have demonstrated promising efficiency by aggressively reducing parameter counts and computational complexity. However, these models often struggle to accurately delineate fine lesion boundaries and texture patterns essential for early skin cancer diagnosis and treatment planning. In this paper, we propose MambaLiteUNet, a compact yet robust segmentation framework that integrates Mamba state space modeling into a U-Net architecture, along with three key modules: Adaptive Multi-Branch Mamba Feature Fusion (AMF), Local-Global Feature Mixing (LGFM), and Cross-Gated Attention (CGA). These modules are designed to enhance local-global feature interaction, preserve spatial details, and improve the quality of skip connections. MambaLiteUNet achieves an average IoU of 87.12% and average Dice score of 93.09% across ISIC2017, ISIC2018, HAM10000, and PH2 benchmarks, outperforming state-of-the-art models. Compared to U-Net, our model improves average IoU and Dice by 7.72 and 4.61 points, respectively, while reducing parameters by 93.6% and GFLOPs by 97.6%. Additionally, in domain generalization with six unseen lesion categories, MambaLiteUNet achieves 77.61% IoU and 87.23% Dice, performing best among all evaluated models. Our extensive experiments demonstrate that MambaLiteUNet achieves a strong balance between accuracy and efficiency, making it a competitive and practical solution for dermatological image segmentation. Our code is publicly available at: https://github.com/maklachur/MambaLiteUNet.

Problem

• skin lesion segmentation
• fine boundary delineation
• texture pattern preservation
• early skin cancer diagnosis
• medical image segmentation

Innovation

• Mamba state space model
• Cross-Gated Attention
• Adaptive Feature Fusion
• Lightweight segmentation
• Skin lesion segmentation