Convolution Neural network based Marine Predator Algorithm for COVID-19 detection

Abstract

COVID-19 is among the dangerous illness in the world due to its quickly spreading, posing a new challenge for researchers to discover it early. In last few months, new covid19 virus strains have been found in South Africa, India, and United Kingdom (UK) due to the mutation of the virus. Owing this critical situation of the world health and with increased number of the cases with the absence of efficient a cure vaccine, timely quarantine and medical treatment, as well as reliable identification of COVID-19, are required to prevent and contain this pandemic. Radiology images and Artificial Intelligence techniques are the most used techniques in computer-aided medical diagnosis for Covid-19 detection. The present paper shows a Convolution Neural Network based novel metaheuristic techniques called Marine Predator Algorithm for detecting Covid-19 and well differentiate between Covid-19 and Pneumonia disease. Our proposed system achieves good results in term of classification such as 93% of accuracy, 95% of precision, 97% of recall and F1-score 95%.