Automatic detection of covid-19 using CNN model combined with Firefly algorithm

Abstract

Coronavirus has already been spread around the world, in many countries, and it has already claimed many lives. Further, the World Health Organization (WHO) has notified public health officials that COVID-19 has reached global epidemic status. Therefore, an early diagnosis using a chest CT scan can aid medical specialists in critical situations. This study aims to develop a web-based service for detecting COVID-19 online. To achieve our goal, we merged the convolutional neural network (CNN) model with the Firefly algorithm (FA). This combination ameliorate definitely the performance and efficiency of the CNN proposed model. Furthermore, the experiments revealed that the proposed FACNN framework enables us to reach high performance with regard to precision, accuracy, sensitivity, F-measure, recall and specificity (1.0%, 1.0%, 1.0%, 1.0%, 1.0% and 1.0%). In addition, a web-based interface was developed to identify and recogonize COVID-19 in chest radiographs in just few seconds. We anticipate that this web predictor will potentially save precious lives, and therefore contribute to society positively.