SkinMamba: A Precision Skin Lesion Segmentation Architecture with Cross-Scale Global State Modeling and Frequency Boundary Guidance

Shun Zou, Mingya Zhang, Bingjian Fan, Zhengyi Zhou, Xiuguo Zou
{"title":"SkinMamba: A Precision Skin Lesion Segmentation Architecture with Cross-Scale Global State Modeling and Frequency Boundary Guidance","authors":"Shun Zou, Mingya Zhang, Bingjian Fan, Zhengyi Zhou, Xiuguo Zou","doi":"arxiv-2409.10890","DOIUrl":null,"url":null,"abstract":"Skin lesion segmentation is a crucial method for identifying early skin\ncancer. In recent years, both convolutional neural network (CNN) and\nTransformer-based methods have been widely applied. Moreover, combining CNN and\nTransformer effectively integrates global and local relationships, but remains\nlimited by the quadratic complexity of Transformer. To address this, we propose\na hybrid architecture based on Mamba and CNN, called SkinMamba. It maintains\nlinear complexity while offering powerful long-range dependency modeling and\nlocal feature extraction capabilities. Specifically, we introduce the Scale\nResidual State Space Block (SRSSB), which captures global contextual\nrelationships and cross-scale information exchange at a macro level, enabling\nexpert communication in a global state. This effectively addresses challenges\nin skin lesion segmentation related to varying lesion sizes and inconspicuous\ntarget areas. Additionally, to mitigate boundary blurring and information loss\nduring model downsampling, we introduce the Frequency Boundary Guided Module\n(FBGM), providing sufficient boundary priors to guide precise boundary\nsegmentation, while also using the retained information to assist the decoder\nin the decoding process. Finally, we conducted comparative and ablation\nexperiments on two public lesion segmentation datasets (ISIC2017 and ISIC2018),\nand the results demonstrate the strong competitiveness of SkinMamba in skin\nlesion segmentation tasks. The code is available at\nhttps://github.com/zs1314/SkinMamba.","PeriodicalId":501289,"journal":{"name":"arXiv - EE - Image and Video Processing","volume":null,"pages":null},"PeriodicalIF":0.0000,"publicationDate":"2024-09-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"arXiv - EE - Image and Video Processing","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/arxiv-2409.10890","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
引用次数: 0

Abstract

Skin lesion segmentation is a crucial method for identifying early skin cancer. In recent years, both convolutional neural network (CNN) and Transformer-based methods have been widely applied. Moreover, combining CNN and Transformer effectively integrates global and local relationships, but remains limited by the quadratic complexity of Transformer. To address this, we propose a hybrid architecture based on Mamba and CNN, called SkinMamba. It maintains linear complexity while offering powerful long-range dependency modeling and local feature extraction capabilities. Specifically, we introduce the Scale Residual State Space Block (SRSSB), which captures global contextual relationships and cross-scale information exchange at a macro level, enabling expert communication in a global state. This effectively addresses challenges in skin lesion segmentation related to varying lesion sizes and inconspicuous target areas. Additionally, to mitigate boundary blurring and information loss during model downsampling, we introduce the Frequency Boundary Guided Module (FBGM), providing sufficient boundary priors to guide precise boundary segmentation, while also using the retained information to assist the decoder in the decoding process. Finally, we conducted comparative and ablation experiments on two public lesion segmentation datasets (ISIC2017 and ISIC2018), and the results demonstrate the strong competitiveness of SkinMamba in skin lesion segmentation tasks. The code is available at https://github.com/zs1314/SkinMamba.
SkinMamba:具有跨尺度全球状态建模和频率边界指导功能的精确皮肤病变分割架构
皮损分割是识别早期皮肤癌的重要方法。近年来,基于卷积神经网络(CNN)和变换器的方法得到了广泛应用。此外,将卷积神经网络和变换器相结合能有效整合全局和局部关系,但仍受限于变换器的二次复杂性。为此,我们提出了一种基于 Mamba 和 CNN 的混合架构,称为 SkinMamba。它在保持线性复杂性的同时,提供了强大的长距离依赖建模和局部特征提取功能。具体来说,我们引入了尺度残留状态空间块(SRSSB),它能在宏观层面捕捉全局上下文关系和跨尺度信息交换,实现全局状态下的专家交流。这有效解决了皮损大小不一和目标区域不明显等皮损分割难题。此外,为了减少模型下采样时的边界模糊和信息丢失,我们引入了频率边界引导模块(FBGM),提供足够的边界先验来引导精确的边界分割,同时还利用保留的信息来协助解码器进行解码。最后,我们在两个公开的皮损分割数据集(ISIC2017 和 ISIC2018)上进行了对比和消融实验,结果表明 SkinMamba 在皮损分割任务中具有很强的竞争力。代码可在https://github.com/zs1314/SkinMamba。
本文章由计算机程序翻译,如有差异,请以英文原文为准。
求助全文
约1分钟内获得全文 求助全文
来源期刊
自引率
0.00%
发文量
0
×
引用
GB/T 7714-2015
复制
MLA
复制
APA
复制
导出至
BibTeX EndNote RefMan NoteFirst NoteExpress
×
提示
您的信息不完整,为了账户安全,请先补充。
现在去补充
×
提示
您因"违规操作"
具体请查看互助需知
我知道了
×
提示
确定
请完成安全验证×
copy
已复制链接
快去分享给好友吧!
我知道了
右上角分享
点击右上角分享
0
联系我们:info@booksci.cn Book学术提供免费学术资源搜索服务,方便国内外学者检索中英文文献。致力于提供最便捷和优质的服务体验。 Copyright © 2023 布克学术 All rights reserved.
京ICP备2023020795号-1
ghs 京公网安备 11010802042870号
Book学术文献互助
Book学术文献互助群
群 号:481959085
Book学术官方微信