Classification of COVID-19 Cases using Deep Neural Network based on Chest Image Data through WSN

Due to the rapid spread of corona virus disease (COVID-19), it has been considered as a pandemic throughout the world. The misclassification of COVID-19 cases may even lead the death of the patients, and hence the diagnosis at early stage is important to stop further spread of the infection and to safeguard the life of the patients. This paper proposes the Aquila tuned Deep neural network (Aquila-DNN) classifier for the classification of COVID-19 patients using the chest image data assessed through Wireless sensor Network (WSN). The extraction of important features from the chest image data is important in the diagnosis as it encloses the important data of the patients. The optimal tuning of the DNN parameters using the Aquila Optimizer (AO) assists in improving the classification accuracy of proposed model. In addition, the convergence is also boosted using the tuning process of the AO algorithm. The effectiveness of the proposed Aquila-DNN model is validated with the analysis of the model based on the performance indices, namely accuracy, ROC curve, and F1 measure. The testing accuracy and the training accuracy of Aquila-DNN model are attained to be 99.7%, and 95.4545%, respectively. © 2022, Ismail Saritas. All rights reserved.