深度学习在肺炎检测中的研究综述  

Review of deep learning in pneumonia detection

作　　者：李新[1,2] 陈帆 郝海江 黄琳[1,2] 刘亚荣[1,2] LI Xin;CHEN Fan;HAO Hai-jiang;HUANG Lin;LIU Ya-rong(School of Information Science and Engineering,Guilin University of Technology,Guilin 541006,China;Guangxi Key Laboratory of Embedded Technology and Intelligent Systems,Guilin University of Technology,Guilin 541006,China)

机构地区：[1]桂林理工大学信息科学与工程学院,广西桂林541006 [2]桂林理工大学广西嵌入式技术与智能系统重点实验室,广西桂林541006

出　　处：《桂林理工大学学报》2020年第4期859-866,共8页Journal of Guilin University of Technology

基　　金：广西自然科学基金项目(2018GXNSFBA281081);广西嵌入式技术与智能系统重点实验室开放基金项目(RZ18103089,2018A-10,RZ2000001651,2019-02-03)。

摘　　要：针对专业影像医生每天因阅览大量胸片所导致的视觉疲劳,会在一定程度上影响医生的诊断,从而出现假阳率过高、误诊漏诊等问题。为了减少这类情况的出现,近年来大量研究将深度学习技术应用于医学影像辅助诊断方面以减轻医生负担、提高医院检测工作效率。本文对比了深度学习中几种常用的卷积神经网络模型,并且总结了几个比较大型的肺炎公开数据集,然后以肺炎检测为综述对象,从传统方法和深度学习方法两个方面和分类、目标检测两个角度,详细分析了近年来国内外的研究现状、取得成果以及所面临的问题,并给出相关建议。To reduce the visual fatigue of professional imaging doctors in reading a large number of chest radiographs and the doctors diagnosis,high positive rate,and good diagnosis are expected.Most scholars begin to apply deep learning technology to medical imaging,so as to reduce the burden of doctors and improve the efficiency of hospital testing.Several convolutional neural network models commonly used in deep learning were compared.And several large public data sets of pneumonia are summarized.Take pneumonia detection as the review object,traditional methods and deep learning methods are analyzed in detail.The research status and review deep learning in pneumonia detection at home and abroad are summarized,and some suggestions are given in this paper.

关 键 词：深度学习 医学影像 肺炎检测 计算机辅助诊断 

分 类 号：TP391.41[自动化与计算机技术—计算机应用技术]