UltraLBM-UNet: Ultralight Bidirectional Mamba-based Model for Skin Lesion Segmentation

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Fan, Linxuan, Jiang, Juntao, Liu, Weixuan, Xue, Zhucun, Lv, Jiajun, Zhang, Jiangning, Liu, Yong
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866908731854290944
author Fan, Linxuan
Jiang, Juntao
Liu, Weixuan
Xue, Zhucun
Lv, Jiajun
Zhang, Jiangning
Liu, Yong
author_facet Fan, Linxuan
Jiang, Juntao
Liu, Weixuan
Xue, Zhucun
Lv, Jiajun
Zhang, Jiangning
Liu, Yong
contents 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.
format Preprint
id arxiv_https___arxiv_org_abs_2512_21584
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle UltraLBM-UNet: Ultralight Bidirectional Mamba-based Model for Skin Lesion Segmentation
Fan, Linxuan
Jiang, Juntao
Liu, Weixuan
Xue, Zhucun
Lv, Jiajun
Zhang, Jiangning
Liu, Yong
Computer Vision and Pattern Recognition
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.
title UltraLBM-UNet: Ultralight Bidirectional Mamba-based Model for Skin Lesion Segmentation
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2512.21584