A new approach to predict COVID-19 using artificial neural networks

In COVID-19, most of the patients have been diagnosed with pneumonia in their early stages. Most of the symptoms that have been in the display or have evolved in the last couple of months like fever, cough, and shortness of breath have been predominant. Moreover, based on the studies and reports of the infected patients, symptoms like heart disease, hypertension, chest pain, diarrhea, and nasal congestion have shown a significant impact in the sustenance of COVID-19. Taking all these symptoms into consideration along with the person’s age, a prediction process has been developed in this chapter to check whether the person is infected with COVID-19 or not. Based on the significance of these attributes, we have applied artificial neural network to classify the patient’s condition into three classes, which include no infection, mild infection, and serious infection. We have achieved an accuracy of 84.7% in predicting the cases.