Detecting COVID-19 in X-ray Images using Transfer Learning

Authors

Jamal Alsakran, Computer Information Science, Higher Colleges of Technology, Fujairah, UAEFollow
Loai Alnemer, Department of Computer Information Science, King Abulllah II School for Information Technology, The University of Jordan, Amman, JordanFollow
Nouh Alhindawi, Department of Software Engineering, Faculty of Sciences and Information Technology, Jadara University, Irbid, JordanFollow
Omayya Muard, Department of Research and Development, Association of Arab Universities, Amman, JordanFollow

Abstract

Accurate and speedy detection of COVID-19 is essential to curb the spread of the disease and avoid overwhelming the health care system. COVID-19 detection using X-ray images is commonly practiced at medical centers; however, it requires the intervention of medical professionals trained in diagnosing and interpreting medical imagining. In this paper, we employ deep transfer learning models to detect COVID-19 on a dataset of over 20,000 X-ray images. Our results on 5 pretrained models (VGG19, InceptionV3, MobileNetV2, DenseNet121, and ResNet101V2) show high performance of 99% without image augmentation, and 93\% when image augmentation is used.

Recommended Citation

Alsakran, Jamal; Alnemer, Loai; Alhindawi, Nouh; and Muard, Omayya (2022) "Detecting COVID-19 in X-ray Images using Transfer Learning," Information Sciences Letters: Vol. 11 : Iss. 5 , PP -.
Available at: https://digitalcommons.aaru.edu.jo/isl/vol11/iss5/38