Classification of COVID-19 Cases: An Exploratory Study by Incorporating Transfer Learning with Cloud

Authors

Piyush Vyas, Dakota State UniversityFollow
Kaushik Ragothaman, Dakota State UniversityFollow
Akhilesh Chauhan, Dakota State UniversityFollow
Bhaskar Rimal, Dakota State UniversityFollow

Abstract

The coronavirus pandemic raised several challenges globally due to heavy demand for patient care and led researchers to find various methods to detect coronavirus. This work aims to provide a transfer learning (TL) based approach for detecting the COVID-19 cases by employing cloud computing thereby minimizing the processing time and costs. In contrast to the previous studies, we have used real-time COVID-19 positive cases chest X-ray images for the training of adopted pre-trained models such as VGG16, InceptionV3, and DenseNet121. The obtained results showed that VGG16 outperforms those models by 98.21% to precisely classify the COVID-19 positive and negative cases.

Recommended Citation

Vyas, Piyush; Ragothaman, Kaushik; Chauhan, Akhilesh; and Rimal, Bhaskar, "Classification of COVID-19 Cases: An Exploratory Study by Incorporating Transfer Learning with Cloud" (2021). MWAIS 2021 Proceedings. 8.
https://aisel.aisnet.org/mwais2021/8