The paper introduces LightM-UNet, a lightweight network architecture that integrates the Mamba State Space Model (SSM) into the UNet framework for medical image segmentation. UNet, a widely used model, faces challenges due to its large parameter count and computational demands, making it unsuitable for mobile health applications. Mamba, a lightweight SSM, addresses these issues by providing linear computational complexity and efficient long-range spatial dependency modeling. LightM-UNet leverages the Residual Vision Mamba Layer (RVM Layer) to extract deep semantic features and model long-range spatial dependencies with minimal additional parameters and computational overhead. Extensive experiments on 2D and 3D datasets, including LiTs and Montgomery&Shenzhen, demonstrate that LightM-UNet outperforms existing state-of-the-art models, achieving superior segmentation performance while reducing parameters and computational costs by 116x and 21x, respectively, compared to nnU-Net. The paper also includes ablation studies to validate the effectiveness of the proposed modules and concludes with future directions for further lightweight network design and broader dataset validation.The paper introduces LightM-UNet, a lightweight network architecture that integrates the Mamba State Space Model (SSM) into the UNet framework for medical image segmentation. UNet, a widely used model, faces challenges due to its large parameter count and computational demands, making it unsuitable for mobile health applications. Mamba, a lightweight SSM, addresses these issues by providing linear computational complexity and efficient long-range spatial dependency modeling. LightM-UNet leverages the Residual Vision Mamba Layer (RVM Layer) to extract deep semantic features and model long-range spatial dependencies with minimal additional parameters and computational overhead. Extensive experiments on 2D and 3D datasets, including LiTs and Montgomery&Shenzhen, demonstrate that LightM-UNet outperforms existing state-of-the-art models, achieving superior segmentation performance while reducing parameters and computational costs by 116x and 21x, respectively, compared to nnU-Net. The paper also includes ablation studies to validate the effectiveness of the proposed modules and concludes with future directions for further lightweight network design and broader dataset validation.