UltraLBM-UNet: Ultralight Bidirectional Mamba-based Model for Skin Lesion Segmentation
Preprint
Abstract
Skin lesion segmentation is a crucial step in dermatology for guiding clinical decision-making. However, existing methods for accurate, robust, and resource-efficient lesion analysis have limitations, including low performance and high computational complexity. To address these limitations, we propose UltraLBM-UNet, a lightweight U-Net variant that integrates a bidirectional Mamba-based global modeling mechanism with multi-branch local feature perception. The proposed architecture integrates efficient local feature injection with bidirectional state-space modeling, enabling richer contextual interaction across spatial dimensions while maintaining computational compactness suitable for point-of-care deployment. Extensive experiments on the ISIC 2017, ISIC 2018, and PH2 datasets demonstrate that our model consistently achieves state-of-the-art segmentation accuracy, outperforming existing lightweight and Mamba counterparts with only 0.034M parameters and 0.060 GFLOPs. In addition, we introduce a hybrid knowledge distillation strategy to train an ultra-compact student model, where the distilled variant UltraLBM-UNet-T, with only 0.011M parameters and 0.019 GFLOPs, achieves competitive segmentation performance. These results highlight the suitability of UltraLBM-UNet for point-of-care deployment, where accurate and robust lesion analyses are essential. The source code is publicly available at https://github.com/LinLinLin-X/UltraLBM-UNet.Keywords
All Related Versions
This publication has 15 references indexed in Scilit:
- UltraLight VM-UNet: Parallel Vision Mamba significantly reduces parameters for skin lesion segmentationPatterns, 2025
- DSEUNet: A lightweight UNet for dynamic space grouping enhancement for skin lesion segmentationExpert Systems with Applications, 2024
- LeaNet: Lightweight U-shaped architecture for high-performance skin cancer image segmentationComputers in Biology and Medicine, 2024
- Lite-UNet: A lightweight and efficient network for cell localizationEngineering Applications of Artificial Intelligence, 2023
- LMU-Net: lightweight U-shaped network for medical image segmentationMedical & Biological Engineering & Computing, 2023
- A survey on deep learning for skin lesion segmentationMedical Image Analysis, 2023
- Swin-Unet: Unet-Like Pure Transformer for Medical Image Segmentation2023
- The HAM10000 dataset, a large collection of multi-source dermatoscopic images of common pigmented skin lesionsScientific Data, 2018
- Dermatologist-level classification of skin cancer with deep neural networksNature, 2017
- V-Net: Fully Convolutional Neural Networks for Volumetric Medical Image Segmentation2016