Rapid detection of COVID-19 from chest X-ray images using deep convolutional neural networks

IF 0.7 Q4 ENGINEERING, BIOMEDICAL

International Journal of Biomedical Engineering and Technology Pub Date : 2023-01-01 DOI:10.1504/ijbet.2023.10053528

K. V. Kadambari, S. Panigrahi, U. Raju, Harika Ala, Debanjan Pathak

{"title":"Rapid detection of COVID-19 from chest X-ray images using deep convolutional neural networks","authors":"K. V. Kadambari, S. Panigrahi, U. Raju, Harika Ala, Debanjan Pathak","doi":"10.1504/ijbet.2023.10053528","DOIUrl":null,"url":null,"abstract":"The entire world is suffering from the corona pandemic (COVID-19) since December 2019. Deep convolutional neural networks (deep CNN) can be used to develop a rapid detection system of COVID-19. Among all the existing literature, ResNet50 is showing better performance, but with three main limitations, i.e.: 1) overfitting;2) computation cost;3) loss of feature information. To overcome these problems authors have proposed four different modifications on ResNet50, naming it as LightWeightResNet50. An image dataset containing chest X-ray images of coronavirus patients and normal persons is used for evaluation. Five-fold cross-validation is applied with transfer learning. Ten different performance measures (true positive, false negative, false positive, true negative, accuracy, recall, specificity, precision, F1-score and area under curve) are used for evaluation along with fold-wise performance measures comparison. The four proposed methods have an accuracy improvement of 4%, 13%, 14% and 7% respectively when compared with ResNet50.","PeriodicalId":51752,"journal":{"name":"International Journal of Biomedical Engineering and Technology","volume":"1 1","pages":""},"PeriodicalIF":0.7000,"publicationDate":"2023-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"International Journal of Biomedical Engineering and Technology","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1504/ijbet.2023.10053528","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q4","JCRName":"ENGINEERING, BIOMEDICAL","Score":null,"Total":0}

引用次数: 0

Abstract

The entire world is suffering from the corona pandemic (COVID-19) since December 2019. Deep convolutional neural networks (deep CNN) can be used to develop a rapid detection system of COVID-19. Among all the existing literature, ResNet50 is showing better performance, but with three main limitations, i.e.: 1) overfitting;2) computation cost;3) loss of feature information. To overcome these problems authors have proposed four different modifications on ResNet50, naming it as LightWeightResNet50. An image dataset containing chest X-ray images of coronavirus patients and normal persons is used for evaluation. Five-fold cross-validation is applied with transfer learning. Ten different performance measures (true positive, false negative, false positive, true negative, accuracy, recall, specificity, precision, F1-score and area under curve) are used for evaluation along with fold-wise performance measures comparison. The four proposed methods have an accuracy improvement of 4%, 13%, 14% and 7% respectively when compared with ResNet50.

查看原文本刊更多论文

基于深度卷积神经网络的胸部x线图像快速检测COVID-19

自2019年12月以来，全世界都在遭受冠状病毒大流行(COVID-19)的折磨。深度卷积神经网络(Deep CNN)可以用于开发COVID-19的快速检测系统。在现有的文献中，ResNet50表现出更好的性能，但主要存在三个局限性，即:1)过拟合;2)计算成本;3)特征信息丢失。为了克服这些问题，作者对ResNet50提出了四种不同的修改，将其命名为LightWeightResNet50。使用包含冠状病毒患者和正常人胸部x射线图像的图像数据集进行评估。五重交叉验证应用于迁移学习。采用10种不同的性能指标(真阳性、假阴性、假阳性、真阴性、准确性、召回率、特异性、精密度、f1评分和曲线下面积)进行评估，并对各指标进行比较。与ResNet50相比，这四种方法的准确率分别提高了4%、13%、14%和7%。

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

International Journal of Biomedical Engineering and Technology ENGINEERING, BIOMEDICAL-

CiteScore

1.60

自引率

0.00%

发文量

期刊介绍： IJBET addresses cutting-edge research in the multi-disciplinary area of biomedical engineering and technology. Medical science incorporates scientific/technological advances combining to produce more accurate diagnoses, effective treatments with fewer side effects, and improved ability to prevent disease and provide superior-quality healthcare. A key field here is biomedical engineering/technology, offering a synthesis of physical, chemical, mathematical and computational sciences combined with engineering principles to enhance R&D in biology, medicine, behaviour, and health. Topics covered include Artificial organs Automated patient monitoring Advanced therapeutic and surgical devices Application of expert systems and AI to clinical decision making Biomaterials design Biomechanics of injury and wound healing Blood chemistry sensors Computer modelling of physiologic systems Design of optimal clinical laboratories Medical imaging systems Sports medicine.