CT image classification of COVID-19 based on VGG16 image classification algorithm

Dan Wang; Qinyu Zhao; Guoyi Xuan; Haiyan Zhang

doi:10.54254/2753-8818/32/20240838

Dan Wang , Qinyu Zhao , Guoyi Xuan , Haiyan Zhang^*

Shandong University of Traditional Chinese Medicine

^* Corresponding author: zhy2013@sdutcm.edu.cn

Published: 6 March 2024

DOI: https://doi.org/10.54254/2753-8818/32/20240838

Abstract

This paper introduces the background of the novel coronavirus outbreak and the role of CT scanning in diagnosis, as well as the importance of automated CT image analysis system. COVID-19, a highly contagious respiratory disease caused by the novel coronavirus, has caused millions of infections and hundreds of thousands of deaths worldwide since it broke out in Wuhan, China, in early 2020. CT scan is a commonly used diagnostic means to help doctors determine the extent and location of lung lesions, but due to the large number of COVID-19 patients, doctors have a large workload, so automated CT image analysis system is needed to assist doctors in rapid diagnosis. With the continuous training process, the accuracy of the training set of automated CT image analysis system is gradually stabilized at 90%. The accuracy of the test set exceeds 85%, and the prediction effect is good. The loss of training set and test set became smaller and the training accuracy gradually improved. According to the confusion matrix, 6 CT images that should have been COVID-19 were predicted to be non-COVID-19, 57 CT images that should have been non-COVID-19 were predicted to be COVID-19, and the remaining images were predicted correctly, and the prediction effect of the model was good, which could predict the COVID-19 images more accurately. The application of automated CT image analysis system can greatly reduce the work burden of doctors, improve the efficiency and accuracy of diagnosis, and provide strong technical support for the prevention, control and treatment of COVID-19. At the same time, the continuous optimization and improvement of the system will also provide more effective technical means for future epidemic prevention and control and medical diagnosis. However, there are still some problems and challenges in the application of automated CT image analysis system. For example, there may be bias in the model’s training data, resulting in inaccurate model predictions. At the same time, the case data in different regions are quite different, so targeted training and optimization are needed. Therefore, it is necessary to continuously improve and optimize the system to improve its prediction effect and reliability. In conclusion, the application of automated CT image analysis system provides strong technical support for the prevention, control and treatment of the novel coronavirus pneumonia, which can quickly and accurately diagnose lung lesions and provide better treatment and care for patients. With the continuous development and improvement of the technology, the application prospect of the system will be broader and make greater contributions to the cause of human health.

Keywords: VGG16 , COVID-19 , Accuracy

TNS Vol.32

ISSN (Print): 2753-8818

ISSN (Online): 2753-8826

ISBN (Print): 978-1-83558-321-0

ISBN (Online): 978-1-83558-322-7

Download Cover

Article Information

Received: 27 November 2023

Published: 6 March 2024

DOI: https://doi.org/10.54254/2753-8818/32/20240838

Article Type: Research Article

Cite this article

Wang,D.;Zhao,Q.;Xuan,G.;Zhang,H. (2024). CT image classification of COVID-19 based on VGG16 image classification algorithm. Theoretical and Natural Science,32,177-183.

Data availability

The datasets used and/or analyzed during the current study will be available from the authors upon reasonable request.

Disclaimer/Publisher's Note

The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of EWA Publishing and/or the editor(s). EWA Publishing and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content.

About this Volume

Volume Title: TNS Vol.32

Part of Series: Theoretical and Natural Science

ISSN: 2753-8818 (Print) / 2753-8826 (Online)

Article Information

Cite this article

Data availability

Disclaimer/Publisher's Note

About this Volume

Copyright & License