基于CT图像卷积神经网络集成迁移学习模型的COVID-19患者严重程度自动诊断

IF 0.7 Q4 RADIOLOGY, NUCLEAR MEDICINE & MEDICAL IMAGING
Ahmad Shalbaf, P. Gifani, G. Mehri-Kakavand, Mohamad Pursamimi, M. Ghorbani, A. Davanloo, Majid Vafaeezadeh
{"title":"基于CT图像卷积神经网络集成迁移学习模型的COVID-19患者严重程度自动诊断","authors":"Ahmad Shalbaf, P. Gifani, G. Mehri-Kakavand, Mohamad Pursamimi, M. Ghorbani, A. Davanloo, Majid Vafaeezadeh","doi":"10.2478/pjmpe-2022-0014","DOIUrl":null,"url":null,"abstract":"Abstract Introduction: Quantification of lung involvement in COVID-19 using chest Computed tomography (CT) scan can help physicians to evaluate the progression of the disease or treatment response. This paper presents an automatic deep transfer learning ensemble based on pre-trained convolutional neural networks (CNNs) to determine the severity of COVID -19 as normal, mild, moderate, and severe based on the images of the lungs CT. Material and methods: In this study, two different deep transfer learning strategies were used. In the first procedure, features were extracted from fifteen pre-trained CNNs architectures and then fed into a support vector machine (SVM) classifier. In the second procedure, the pre-trained CNNs were fine-tuned using the chest CT images, and then features were extracted for the purpose of classification by the softmax layer. Finally, an ensemble method was developed based on majority voting of the deep learning outputs to increase the performance of the recognition on each of the two strategies. A dataset of CT scans was collected and then labeled as normal (314), mild (262), moderate (72), and severe (35) for COVID-19 by the consensus of two highly qualified radiologists. Results: The ensemble of five deep transfer learning outputs named EfficientNetB3, EfficientNetB4, InceptionV3, NasNetMobile, and ResNext50 in the second strategy has better results than the first strategy and also the individual deep transfer learning models in diagnosing the severity of COVID-19 with 85% accuracy. Conclusions: Our proposed study is well suited for quantifying lung involvement of COVID-19 and can help physicians to monitor the progression of the disease.","PeriodicalId":53955,"journal":{"name":"Polish Journal of Medical Physics and Engineering","volume":"17 1","pages":"117 - 126"},"PeriodicalIF":0.7000,"publicationDate":"2022-07-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":"{\"title\":\"Automatic diagnosis of severity of COVID-19 patients using an ensemble of transfer learning models with convolutional neural networks in CT images\",\"authors\":\"Ahmad Shalbaf, P. Gifani, G. Mehri-Kakavand, Mohamad Pursamimi, M. Ghorbani, A. Davanloo, Majid Vafaeezadeh\",\"doi\":\"10.2478/pjmpe-2022-0014\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"Abstract Introduction: Quantification of lung involvement in COVID-19 using chest Computed tomography (CT) scan can help physicians to evaluate the progression of the disease or treatment response. This paper presents an automatic deep transfer learning ensemble based on pre-trained convolutional neural networks (CNNs) to determine the severity of COVID -19 as normal, mild, moderate, and severe based on the images of the lungs CT. Material and methods: In this study, two different deep transfer learning strategies were used. In the first procedure, features were extracted from fifteen pre-trained CNNs architectures and then fed into a support vector machine (SVM) classifier. In the second procedure, the pre-trained CNNs were fine-tuned using the chest CT images, and then features were extracted for the purpose of classification by the softmax layer. Finally, an ensemble method was developed based on majority voting of the deep learning outputs to increase the performance of the recognition on each of the two strategies. A dataset of CT scans was collected and then labeled as normal (314), mild (262), moderate (72), and severe (35) for COVID-19 by the consensus of two highly qualified radiologists. Results: The ensemble of five deep transfer learning outputs named EfficientNetB3, EfficientNetB4, InceptionV3, NasNetMobile, and ResNext50 in the second strategy has better results than the first strategy and also the individual deep transfer learning models in diagnosing the severity of COVID-19 with 85% accuracy. Conclusions: Our proposed study is well suited for quantifying lung involvement of COVID-19 and can help physicians to monitor the progression of the disease.\",\"PeriodicalId\":53955,\"journal\":{\"name\":\"Polish Journal of Medical Physics and Engineering\",\"volume\":\"17 1\",\"pages\":\"117 - 126\"},\"PeriodicalIF\":0.7000,\"publicationDate\":\"2022-07-28\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"\",\"citationCount\":\"0\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"Polish Journal of Medical Physics and Engineering\",\"FirstCategoryId\":\"1085\",\"ListUrlMain\":\"https://doi.org/10.2478/pjmpe-2022-0014\",\"RegionNum\":0,\"RegionCategory\":null,\"ArticlePicture\":[],\"TitleCN\":null,\"AbstractTextCN\":null,\"PMCID\":null,\"EPubDate\":\"\",\"PubModel\":\"\",\"JCR\":\"Q4\",\"JCRName\":\"RADIOLOGY, NUCLEAR MEDICINE & MEDICAL IMAGING\",\"Score\":null,\"Total\":0}","platform":"Semanticscholar","paperid":null,"PeriodicalName":"Polish Journal of Medical Physics and Engineering","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.2478/pjmpe-2022-0014","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q4","JCRName":"RADIOLOGY, NUCLEAR MEDICINE & MEDICAL IMAGING","Score":null,"Total":0}
引用次数: 0

摘要

摘要简介:利用胸部计算机断层扫描(CT)量化COVID-19肺部累及可以帮助医生评估疾病进展或治疗反应。本文提出了一种基于预训练卷积神经网络(cnn)的自动深度迁移学习集成,根据肺部CT图像确定COVID -19的严重程度为正常、轻度、中度和严重。材料与方法:本研究采用了两种不同的深度迁移学习策略。在第一个过程中,从15个预训练的cnn架构中提取特征,然后将其输入到支持向量机(SVM)分类器中。第二步,利用胸部CT图像对预训练好的cnn进行微调,然后通过softmax层提取特征进行分类。最后,开发了一种基于深度学习输出的多数投票的集成方法,以提高对两种策略的识别性能。收集CT扫描数据集,然后在两名高素质放射科医生的共识下,将COVID-19标记为正常(314)、轻度(262)、中度(72)和重度(35)。结果:在第二种策略中,效率netb3、效率netb4、InceptionV3、NasNetMobile和ResNext50五个深度迁移学习输出的集合在诊断COVID-19严重程度方面的效果优于第一种策略,并且单个深度迁移学习模型的准确率为85%。结论:我们提出的研究非常适合量化COVID-19的肺部累及,可以帮助医生监测疾病的进展。
本文章由计算机程序翻译,如有差异,请以英文原文为准。
Automatic diagnosis of severity of COVID-19 patients using an ensemble of transfer learning models with convolutional neural networks in CT images
Abstract Introduction: Quantification of lung involvement in COVID-19 using chest Computed tomography (CT) scan can help physicians to evaluate the progression of the disease or treatment response. This paper presents an automatic deep transfer learning ensemble based on pre-trained convolutional neural networks (CNNs) to determine the severity of COVID -19 as normal, mild, moderate, and severe based on the images of the lungs CT. Material and methods: In this study, two different deep transfer learning strategies were used. In the first procedure, features were extracted from fifteen pre-trained CNNs architectures and then fed into a support vector machine (SVM) classifier. In the second procedure, the pre-trained CNNs were fine-tuned using the chest CT images, and then features were extracted for the purpose of classification by the softmax layer. Finally, an ensemble method was developed based on majority voting of the deep learning outputs to increase the performance of the recognition on each of the two strategies. A dataset of CT scans was collected and then labeled as normal (314), mild (262), moderate (72), and severe (35) for COVID-19 by the consensus of two highly qualified radiologists. Results: The ensemble of five deep transfer learning outputs named EfficientNetB3, EfficientNetB4, InceptionV3, NasNetMobile, and ResNext50 in the second strategy has better results than the first strategy and also the individual deep transfer learning models in diagnosing the severity of COVID-19 with 85% accuracy. Conclusions: Our proposed study is well suited for quantifying lung involvement of COVID-19 and can help physicians to monitor the progression of the disease.
求助全文
通过发布文献求助,成功后即可免费获取论文全文。 去求助
来源期刊
Polish Journal of Medical Physics and Engineering
Polish Journal of Medical Physics and Engineering RADIOLOGY, NUCLEAR MEDICINE & MEDICAL IMAGING-
CiteScore
1.30
自引率
0.00%
发文量
19
期刊介绍: Polish Journal of Medical Physics and Engineering (PJMPE) (Online ISSN: 1898-0309; Print ISSN: 1425-4689) is an official publication of the Polish Society of Medical Physics. It is a peer-reviewed, open access scientific journal with no publication fees. The issues are published quarterly online. The Journal publishes original contribution in medical physics and biomedical engineering.
×
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