Covid-19 Chest X-Ray Classification using Convolutional Neural Network

• Published in: 14th IEEE International Conference on Control System, Computing and Engineering, ICCSCE 2024 - Proceedings
• Main Author: Moqbel M.A.A.A.; Radzi M.D.I.M.; Halim N.H.A.; Soh Z.H.C.; Osman M.K.; Saad Z.
• Format: Conference paper
• Language: English
• Published: Institute of Electrical and Electronics Engineers Inc. 2024
• Online Access: https://www.scopus.com/inward/record.uri?eid=2-s2.0-85207102076&doi=10.1109%2fICCSCE61582.2024.10696059&partnerID=40&md5=42235ad135d99f6b98ba47f48c1aa222

COVID-19, or Coronavirus Disease 2019, is an infectious ailment caused by the SARS coronavirus 2 (SARS-CoV-2). Identifying infected individuals poses a challenge due to symptom similarities with other common illnesses like fever and cough. One method to assess infection involves examining chest X-rays, but with the volume of images, misdiagnoses may occur, causing delays. To address this, an automated COVID-19 classification using Convolutional Neural Network (CNN) techniques is proposed. Models like AlexNet, SqueezeNet, MobileNetV2, and ResNet-18 are employed and trained on datasets containing normal and COVID-19 chest X-rays, with efforts to expand and enhance the datasets. In the comparative analysis of pre-trained models for chest X-ray image classification, MobileNetV2 stands out as the superior model, achieving an impressive average accuracy of 96% alongside equivalent precision, recall, and F1-score metrics. This remarkable performance underscores MobileNetV2's efficacy in distinguishing between normal and COVID-19 affected images, positioning it as the most suited model for this crucial diagnostic task. © 2024 IEEE.