Journal
2020 1ST INTERNATIONAL CONFERENCE ON INFORMATION TECHNOLOGY, ADVANCED MECHANICAL AND ELECTRICAL ENGINEERING (ICITAMEE 2020)
Volume -, Issue -, Pages 232-237Publisher
IEEE
DOI: 10.1109/ICITAMEE50454.2020.9398392
Keywords
COVID-19; CXR images; deep learning; diagnosis
Ask authors/readers for more resources
Recently, Corona Virus Disease 2019 (COVID-19) has rapidly spanned the globe. In particular, this viral disease has infected more than 400,000 peoples and has caused more than twenty thousand cases of death. Unfortunately, there is no specific therapeutic drugs or vaccines for the disease, such that an early screening protocol is highly required. Although nucleic acid detection using real-time polymerase chain reaction (RT-PCR) remains the standard, recent literature reported that radiological imaging of human chests had shown a more consistent result when used for COVID-19 diagnosis. However, performing a manual evaluation on chest computed tomography or CXR images is tedious and labour-extensive. In this paper, we present COVID-19Net, a deep neural network-based algorithm to assist doctors in diagnosing COVID-19 through the radiographic images. In the experimental parts, our algorithm could diagnose COVID-19 and other related diseases like SARS, Streptococcus, ARDS, and Pneumocystis with average accuracy and area under the ROC curve (AUC) of > 99% and > 0.99, respectively.
Authors
I am an author on this paper
Click your name to claim this paper and add it to your profile.
Reviews
Recommended
No Data Available