Neural Networks Architecture for COVID-19 Early Detection

Zgheib, Rita
Chahbandarian, Ghazar
Kamalov, Firuz
Labban, Osman El
Journal Title
Journal ISSN
Volume Title
Coronavirus fight seems far from being won. Governments are trying to balance the necessity to enforce restrictions on travel outside the home and the impact of these restrictions on the economy. Healthcare workers are overloaded, a considerable number of unnecessary and costly PCR tests are performed to serve as a certificate to go to work. At this stage, going back to everyday life safely requires the companies and public places to adopt AI-based solutions to assist the public authorities and the hospitals with the COVID detection. The most important issue that we tackle in this paper is the prediction to be very accurate. As a result, we propose an AI system based on Neural Networks (NN) method to predict whether a person has caught COVID19 disease or not. In this study, we used a real data set of 9416 patients tested for COVID19 at a hospital in Dubai. After training the NN model, the average error function of the neural network was equal to 0.01, and the accuracy of the prediction of whether a person has COVID or not was 97.6%. © 2021 IEEE.
Coronavirus, Forecasting, AI systems, Coronaviruses, Government IS, Healthcare workers, Neural network architecture, Neural network method, Neural network model, Public authorities, Public places, Real data sets, Hospitals
Zgheib, R., Chahbandarian, G., Kamalov, F., & Labban, O. E. (2021). Neural networks architecture for COVID-19 early detection. 2021 International Symposium on Networks, Computers and Communications (ISNCC).