MFCPNet: Real time medical image segmentation network via multi-scale feature fusion and channel pruning

IF 4.9 2区 医学 Q1 ENGINEERING, BIOMEDICAL Biomedical Signal Processing and Control Pub Date : 2024-10-29 DOI:10.1016/j.bspc.2024.107074
Linlin Hou , Zishen Yan , Christian Desrosiers , Hui Liu
{"title":"MFCPNet: Real time medical image segmentation network via multi-scale feature fusion and channel pruning","authors":"Linlin Hou ,&nbsp;Zishen Yan ,&nbsp;Christian Desrosiers ,&nbsp;Hui Liu","doi":"10.1016/j.bspc.2024.107074","DOIUrl":null,"url":null,"abstract":"<div><div>Real-time medical image segmentation can not only enhance the interactivity and feasibility of applications but also support more medical application scenarios. Local feature extraction methods reliant on Convolutional Neural Networks (CNN) are hampered by restricted receptive fields, which weakens their ability to capture comprehensive information. Conversely, global feature extraction methods based on Transformers generally face impediments in real-time tasks due to their extensive computational demands. To address these challenges and explore accurate and real-time medical image segmentation models, we introduce this novel MFCPNet. MFCPNet begins by devising Multi-Scale Multi-Channel Convolution (MSMC Conv) to extract local features across various levels and scales. This innovative design contributes to extracting richer local information without unduly burdening the model. Second, for the enhanced receptive field of convolution and the model’s generalization capability, we introduce an Attention Block (Attn Block) carrying rotation invariance. This block, inspired by lightweight Bi-Level Routing Attention (BRA) and MLP-Mixer, effectively mitigates the constraints of convolutional structures and achieves superior contextual modeling. Finally, a judicious pruning of the channel count is employed within MFCPNet, striking a trade-off between segmentation accuracy and efficiency. To evaluate the proposed method, we compare it with several classic approaches using three different types of datasets: retinal images, brain scans, and colon polyps. Across these datasets, MFCPNet achieves segmentation performance comparable to existing methods, with a computational cost of 2.2G FLOPs and 0.49M parameters. Furthermore, it demonstrates a processing speed of 79.54 FPS, meeting the requirements for real-time applications.</div></div>","PeriodicalId":55362,"journal":{"name":"Biomedical Signal Processing and Control","volume":"100 ","pages":"Article 107074"},"PeriodicalIF":4.9000,"publicationDate":"2024-10-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"Biomedical Signal Processing and Control","FirstCategoryId":"5","ListUrlMain":"https://www.sciencedirect.com/science/article/pii/S1746809424011327","RegionNum":2,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q1","JCRName":"ENGINEERING, BIOMEDICAL","Score":null,"Total":0}
引用次数: 0

Abstract

Real-time medical image segmentation can not only enhance the interactivity and feasibility of applications but also support more medical application scenarios. Local feature extraction methods reliant on Convolutional Neural Networks (CNN) are hampered by restricted receptive fields, which weakens their ability to capture comprehensive information. Conversely, global feature extraction methods based on Transformers generally face impediments in real-time tasks due to their extensive computational demands. To address these challenges and explore accurate and real-time medical image segmentation models, we introduce this novel MFCPNet. MFCPNet begins by devising Multi-Scale Multi-Channel Convolution (MSMC Conv) to extract local features across various levels and scales. This innovative design contributes to extracting richer local information without unduly burdening the model. Second, for the enhanced receptive field of convolution and the model’s generalization capability, we introduce an Attention Block (Attn Block) carrying rotation invariance. This block, inspired by lightweight Bi-Level Routing Attention (BRA) and MLP-Mixer, effectively mitigates the constraints of convolutional structures and achieves superior contextual modeling. Finally, a judicious pruning of the channel count is employed within MFCPNet, striking a trade-off between segmentation accuracy and efficiency. To evaluate the proposed method, we compare it with several classic approaches using three different types of datasets: retinal images, brain scans, and colon polyps. Across these datasets, MFCPNet achieves segmentation performance comparable to existing methods, with a computational cost of 2.2G FLOPs and 0.49M parameters. Furthermore, it demonstrates a processing speed of 79.54 FPS, meeting the requirements for real-time applications.
查看原文
分享 分享
微信好友 朋友圈 QQ好友 复制链接
本刊更多论文
MFCPNet:通过多尺度特征融合和通道剪枝实现实时医学图像分割网络
实时医学图像分割不仅能增强应用的交互性和可行性,还能支持更多的医学应用场景。依赖卷积神经网络(CNN)的局部特征提取方法受到感受野的限制,从而削弱了其捕捉综合信息的能力。相反,基于变形器的全局特征提取方法由于计算量大,在实时任务中普遍面临障碍。为了应对这些挑战,探索准确、实时的医学图像分割模型,我们引入了新颖的 MFCPNet。MFCPNet 首先设计了多尺度多通道卷积(MSMC Conv),以提取不同层次和尺度的局部特征。这一创新设计有助于提取更丰富的局部信息,同时又不会给模型带来不必要的负担。其次,为了增强卷积的感受野和模型的泛化能力,我们引入了具有旋转不变性的注意力区块(Attention Block,Attn Block)。受轻量级双层路由注意力(BRA)和 MLP-Mixer 的启发,这个区块有效地减轻了卷积结构的限制,实现了卓越的上下文建模。最后,在 MFCPNet 中采用了明智的通道数修剪方法,在分割准确性和效率之间实现了权衡。为了评估所提出的方法,我们使用三种不同类型的数据集:视网膜图像、大脑扫描和结肠息肉,将其与几种经典方法进行了比较。在这些数据集中,MFCPNet 以 2.2G FLOPs 的计算成本和 0.49M 的参数实现了与现有方法相当的分割性能。此外,它的处理速度达到 79.54 FPS,满足了实时应用的要求。
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 去求助
来源期刊
Biomedical Signal Processing and Control
Biomedical Signal Processing and Control 工程技术-工程:生物医学
CiteScore
9.80
自引率
13.70%
发文量
822
审稿时长
4 months
期刊介绍: Biomedical Signal Processing and Control aims to provide a cross-disciplinary international forum for the interchange of information on research in the measurement and analysis of signals and images in clinical medicine and the biological sciences. Emphasis is placed on contributions dealing with the practical, applications-led research on the use of methods and devices in clinical diagnosis, patient monitoring and management. Biomedical Signal Processing and Control reflects the main areas in which these methods are being used and developed at the interface of both engineering and clinical science. The scope of the journal is defined to include relevant review papers, technical notes, short communications and letters. Tutorial papers and special issues will also be published.
期刊最新文献
PDCA-Net: Parallel dual-channel attention network for polyp segmentation An efficient vision transformer for Alzheimer’s disease classification using magnetic resonance images SLP-Net:An efficient lightweight network for segmentation of skin lesions Automatic segmentation of prostate and organs at risk in CT images using an encoder–decoder structure based on residual neural network A proposal-level class-aware graph convolutional network and memory bank for thyroid nodule detection in ultrasound videos
×
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
现在去查看 取消
×
提示
确定
0
微信
客服QQ
Book学术公众号 扫码关注我们
反馈
×
意见反馈
请填写您的意见或建议
请填写您的手机或邮箱
已复制链接
已复制链接
快去分享给好友吧!
我知道了
×
扫码分享
扫码分享
Book学术官方微信
Book学术文献互助
Book学术文献互助群
群 号:481959085
Book学术
文献互助 智能选刊 最新文献 互助须知 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 Book学术 All rights reserved.
ghs 京公网安备 11010802042870号 京ICP备2023020795号-1