• 文献检索
  • 文档翻译
  • 深度研究
  • 学术资讯
  • Suppr Zotero 插件Zotero 插件
  • 邀请有礼
  • 套餐&价格
  • 历史记录
应用&插件
Suppr Zotero 插件Zotero 插件浏览器插件Mac 客户端Windows 客户端微信小程序
定价
高级版会员购买积分包购买API积分包
服务
文献检索文档翻译深度研究API 文档MCP 服务
关于我们
关于 Suppr公司介绍联系我们用户协议隐私条款
关注我们

Suppr 超能文献

核心技术专利：CN118964589B侵权必究
粤ICP备2023148730 号-1Suppr @ 2026

文献检索

告别复杂PubMed语法，用中文像聊天一样搜索，搜遍4000万医学文献。AI智能推荐，让科研检索更轻松。

立即免费搜索

文件翻译

保留排版，准确专业，支持PDF/Word/PPT等文件格式，支持 12+语言互译。

免费翻译文档

深度研究

AI帮你快速写综述，25分钟生成高质量综述，智能提取关键信息，辅助科研写作。

立即免费体验

用于超声信号分类的轻量级卷积神经网络综述。

A review of lightweight convolutional neural networks for ultrasound signal classification.

作者信息

Zhang Bokun, Li Zhengping, Hao Yuwen, Wang Lijun, Li Xiaoxue, Yao Yuan

机构信息

School of Information Science and Technology, North China University of Technology, Beijing, China.

Disaster Medicine Research Center, Medical Innovation Research Division of the Chinese PLA General Hospital Beijing, China Beijing Key Laboratory of Disaster Medicine, Beijing, China.

出版信息

Front Physiol. 2025 Apr 24;16:1536542. doi: 10.3389/fphys.2025.1536542. eCollection 2025.

DOI:10.3389/fphys.2025.1536542
PMID:40342856
原文链接:https://pmc.ncbi.nlm.nih.gov/articles/PMC12058499/
Abstract

Ultrasound signal processing plays an important role in medical image analysis. Embedded ultrasonography systems with low power consumption and high portability are suitable for disaster rescue, but due to the difficulty of ultrasonic signal recognition, operators need to have strong professional knowledge, and it is not easy to deploy ultrasonography systems in areas with relatively weak infrastructures. In recent years, with the continuous development in the field of deep learning and artificial intelligence, lightweight convolutional neural networks have brought new opportunities for ultrasound signal processing. This paper focuses on investigating lightweight convolutional neural networks applied to ultrasound signal classification. Combined with the characteristics of ultrasound signals, this paper provides a detailed review of lightweight algorithms from two perspectives: model compression and operational optimization. Among them, model compression deals with the overall framework to reduce network redundancy, and the latter aims at the lightweight design of the basic operational module "convolution" in the network. The experimental results of some classical models and algorithms on the ImageNet dataset are summarized. Through the comprehensive analysis, we present some problems and provide an outlook on the future development of lightweight techniques for ultrasound signal classification.

摘要

超声信号处理在医学图像分析中起着重要作用。低功耗、高便携性的嵌入式超声检查系统适用于灾难救援，但由于超声信号识别困难，操作人员需要具备较强的专业知识，并且在基础设施相对薄弱的地区部署超声检查系统并不容易。近年来，随着深度学习和人工智能领域的不断发展，轻量级卷积神经网络为超声信号处理带来了新机遇。本文着重研究应用于超声信号分类的轻量级卷积神经网络。结合超声信号的特点，本文从模型压缩和运算优化两个角度对轻量级算法进行了详细综述。其中，模型压缩针对整体框架以减少网络冗余，后者则针对网络中基本运算模块“卷积”的轻量级设计。总结了一些经典模型和算法在ImageNet数据集上的实验结果。通过综合分析，我们提出了一些问题，并对超声信号分类轻量级技术的未来发展进行了展望。

https://cdn.ncbi.nlm.nih.gov/pmc/blobs/8e43/12058499/8b2716c508a4/fphys-16-1536542-g014.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/8e43/12058499/4f715ba80be7/fphys-16-1536542-g001.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/8e43/12058499/1d1014be5483/fphys-16-1536542-g002.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/8e43/12058499/582d04216f87/fphys-16-1536542-g003.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/8e43/12058499/13918440e774/fphys-16-1536542-g004.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/8e43/12058499/08c8d03cc81c/fphys-16-1536542-g005.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/8e43/12058499/14ff0b2d0cc3/fphys-16-1536542-g006.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/8e43/12058499/6e6439f38af4/fphys-16-1536542-g007.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/8e43/12058499/621fa5c22989/fphys-16-1536542-g008.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/8e43/12058499/55bbd17b62ee/fphys-16-1536542-g009.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/8e43/12058499/4ee423860782/fphys-16-1536542-g010.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/8e43/12058499/d98454a5ff75/fphys-16-1536542-g011.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/8e43/12058499/eebc3407412a/fphys-16-1536542-g012.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/8e43/12058499/9b067be782a9/fphys-16-1536542-g013.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/8e43/12058499/8b2716c508a4/fphys-16-1536542-g014.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/8e43/12058499/4f715ba80be7/fphys-16-1536542-g001.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/8e43/12058499/1d1014be5483/fphys-16-1536542-g002.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/8e43/12058499/582d04216f87/fphys-16-1536542-g003.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/8e43/12058499/13918440e774/fphys-16-1536542-g004.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/8e43/12058499/08c8d03cc81c/fphys-16-1536542-g005.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/8e43/12058499/14ff0b2d0cc3/fphys-16-1536542-g006.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/8e43/12058499/6e6439f38af4/fphys-16-1536542-g007.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/8e43/12058499/621fa5c22989/fphys-16-1536542-g008.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/8e43/12058499/55bbd17b62ee/fphys-16-1536542-g009.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/8e43/12058499/4ee423860782/fphys-16-1536542-g010.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/8e43/12058499/d98454a5ff75/fphys-16-1536542-g011.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/8e43/12058499/eebc3407412a/fphys-16-1536542-g012.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/8e43/12058499/9b067be782a9/fphys-16-1536542-g013.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/8e43/12058499/8b2716c508a4/fphys-16-1536542-g014.jpg

相似文献

1
A review of lightweight convolutional neural networks for ultrasound signal classification.用于超声信号分类的轻量级卷积神经网络综述。
Front Physiol. 2025 Apr 24;16:1536542. doi: 10.3389/fphys.2025.1536542. eCollection 2025.
2
LightMixer: A novel lightweight convolutional neural network for tomato disease detection.LightMixer：一种用于番茄病害检测的新型轻量级卷积神经网络。
Front Plant Sci. 2023 May 9;14:1166296. doi: 10.3389/fpls.2023.1166296. eCollection 2023.
3
Efficient fetal ultrasound image segmentation for automatic head circumference measurement using a lightweight deep convolutional neural network.利用轻量级深度卷积神经网络实现高效胎儿超声图像自动头围测量的分割。
Med Phys. 2022 Aug;49(8):5081-5092. doi: 10.1002/mp.15700. Epub 2022 May 24.
4
[A lightweight convolutional neural network for myositis classification from muscle ultrasound images].[一种用于从肌肉超声图像中进行肌炎分类的轻量级卷积神经网络]
Sheng Wu Yi Xue Gong Cheng Xue Za Zhi. 2024 Oct 25;41(5):895-902. doi: 10.7507/1001-5515.202301023.
5
A Method of Deep Learning Model Optimization for Image Classification on Edge Device.一种用于边缘设备图像分类的深度学习模型优化方法。
Sensors (Basel). 2022 Sep 27;22(19):7344. doi: 10.3390/s22197344.
6
Convolutional neural network in rice disease recognition: accuracy, speed and lightweight.卷积神经网络在水稻病害识别中的应用：准确性、速度与轻量级
Front Plant Sci. 2023 Nov 1;14:1269371. doi: 10.3389/fpls.2023.1269371. eCollection 2023.
7
Research and Application of Ancient Chinese Pattern Restoration Based on Deep Convolutional Neural Network.基于深度卷积神经网络的中国古图案恢复研究与应用。
Comput Intell Neurosci. 2021 Dec 10;2021:2691346. doi: 10.1155/2021/2691346. eCollection 2021.
8
Implementation of Lightweight Convolutional Neural Networks via Layer-Wise Differentiable Compression.通过逐层可微分压缩实现轻量化卷积神经网络。
Sensors (Basel). 2021 May 16;21(10):3464. doi: 10.3390/s21103464.
9
Automatic Modulation Classification Using Hybrid Data Augmentation and Lightweight Neural Network.基于混合数据增强和轻量级神经网络的自动调制分类。
Sensors (Basel). 2023 Apr 22;23(9):4187. doi: 10.3390/s23094187.
10
Focus-RCNet: a lightweight recyclable waste classification algorithm based on focus and knowledge distillation.Focus-RCNet：一种基于注意力机制和知识蒸馏的轻量级可回收垃圾分类算法
Vis Comput Ind Biomed Art. 2023 Oct 11;6(1):19. doi: 10.1186/s42492-023-00146-3.

本文引用的文献

1
Development and Validation of a Class Imbalance-Resilient Cardiac Arrest Prediction Framework Incorporating Multiscale Aggregation, ICA and Explainability.一种融合多尺度聚合、独立成分分析和可解释性的抗类别不平衡心脏骤停预测框架的开发与验证
IEEE Trans Biomed Eng. 2025 May;72(5):1674-1687. doi: 10.1109/TBME.2024.3517635. Epub 2025 Apr 22.
2
Post-Training Network Compression for 3D Medical Image Segmentation: Reducing Computational Efforts via Tucker Decomposition.用于3D医学图像分割的训练后网络压缩：通过塔克分解减少计算量
Radiol Artif Intell. 2025 Mar;7(2):e240353. doi: 10.1148/ryai.240353.
3
Synergistic and Additive Effects of Menadione in Combination with Antibiotics on Multidrug-Resistant Staphylococcus aureus: Insights from Structure-Function Analysis of Naphthoquinones.
甲萘醌与抗生素联合对耐多药金黄色葡萄球菌的协同和相加作用：萘醌结构-功能分析的见解
ChemMedChem. 2023 Dec 14;18(24):e202300328. doi: 10.1002/cmdc.202300328. Epub 2023 Nov 10.
4
3D Ultrasonic Brain Imaging with Deep Learning Based on Fully Convolutional Networks.基于全卷积网络的深度学习三维超声脑成像。
Sensors (Basel). 2023 Oct 9;23(19):8341. doi: 10.3390/s23198341.
5
Machine learning and feature engineering for predicting pulse presence during chest compressions.用于预测胸外按压期间脉搏存在情况的机器学习与特征工程
R Soc Open Sci. 2021 Nov 10;8(11):210566. doi: 10.1098/rsos.210566. eCollection 2021 Nov.
6
Leveraging hallmark Alzheimer's molecular targets using phytoconstituents: Current perspective and emerging trends.利用植物成分针对阿尔茨海默病的标志性分子靶点：当前的观点和新兴趋势。
Biomed Pharmacother. 2021 Jul;139:111634. doi: 10.1016/j.biopha.2021.111634. Epub 2021 May 6.
7
Functional Ultrasound Imaging: A New Imaging Modality for Neuroscience.功能超声成像：神经科学的一种新成像方式。
Neuroscience. 2021 Oct 15;474:110-121. doi: 10.1016/j.neuroscience.2021.03.005. Epub 2021 Mar 13.
8
Prevalence of neural collapse during the terminal phase of deep learning training.深度学习训练末期的神经崩溃的普遍性。
Proc Natl Acad Sci U S A. 2020 Oct 6;117(40):24652-24663. doi: 10.1073/pnas.2015509117. Epub 2020 Sep 21.
9
Addressing class imbalance in deep learning for small lesion detection on medical images.解决医学图像上小病变检测深度学习中的类别不平衡问题。
Comput Biol Med. 2020 May;120:103735. doi: 10.1016/j.compbiomed.2020.103735. Epub 2020 Apr 1.
10
Lattice-Boltzmann interactive blood flow simulation pipeline.晶格玻尔兹曼交互血流模拟流水线。
Int J Comput Assist Radiol Surg. 2020 Apr;15(4):629-639. doi: 10.1007/s11548-020-02120-3. Epub 2020 Mar 4.