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

TL;DR

<3-5 sentence high-level summary> UltraLBM-UNet tackles the need for accurate, robust skin lesion segmentation in resource-constrained, point-of-care settings. It fuses global context via a bidirectional Mamba-based state-space mechanism with local detail through multi-branch perception (GLMBP and LMBP), and introduces scalable skip connections. A hybrid distillation strategy enables an ultra-compact student (UltraLBM-UNet-T) that preserves segmentation quality with dramatically fewer parameters and FLOPs. Evaluations on ISIC 2017/2018 and PH2 demonstrate state-of-the-art performance among lightweight models and strong generalization, highlighting its practical potential for on-device dermatology workflows.

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.

Figures (4)

• Figure 1: Comparative analysis of CNN-, Transformer- and Mamba-based models, highlighting their respective strengths and weaknesses, as well as the superiority of our UltraLBM-UNet.
• Figure 2: Comparison of segmentation models on ISIC2017 and ISIC2018. Each point represents a model, plotted by its parameter size (x-axis) and F1/DSC score (y-axis), with color indicating computational cost in GFLOPs.
• Figure 3: The architecture of UltraLBM-UNet. (a) Overall UltraLBM-UNet Architecture (Encoder-Decoder) with skip-connections and layer dimensions. (b) LMBP Module used in Stage IV with parallel-branch decomposition and residual fusion of local features. (c) GLMBP Module, the core block, integrating Bi-Mamba (global context) and DWConv (local features) via parallel branches. (d) Bi-Mamba Module with shared weights, implemented via sequence flipping for non-causal bi-directional perception.
• Figure 4: Comparison of segmentation results across representative ISIC 2017 cases.