ABSTRACT
An unreported disease cluster occurred in Wuhan, Hubei Province, China in December 2019. The disease was named Coronavirus Disease 2019 (Covid-19) by the World Health Organization (WHO). The transmission of Covid-19 from human to human occurs through coughing/sneezing (drops). Indonesia has Covid-19 cases that are increasing every day. By July 14, 2021, it is counting up to 2,670,046 cases of Covid-19â in Indonesia. This causes Indonesia to become the country with the highest number of Covid-19 cases in Southeast Asia. In dealing with this case, the government has provided various methods to detect the Covid-19 virus in humans, but it has not been able to provide satisfactory results. This is caused by the delay in the detection of the diagnosis given. The delay was due to the limited testing capacity provided. Thus, to help improve patient test results, a CT scan tool is used. The Covid-19 Diagnose System University of Mataram is a system designed to facilitate the process of identifying Covid-19 patients through chest X-rays by applying the CNN model for classification. To find out the system is suitable for the use by end-users, it is necessary to test the usability of the application using the System Usability Scale (SUS) method. The System Usability Scale is used because this method focuses on the end-user. Therefore, this research was made to find out whether the application that has been made (the Covid-19 Diagnose System University of Mataram) can be accepted by users or not. To test the feasibility of this application, it needs to test its usability with the SUS instrument. Testing with SUS has been done using 10 statements of SUS. The test shows the score of SUS is 71.38. It means that the application is acceptable to be used by end users and classified into grade C with a good rating © 2023 Author(s).
ABSTRACT
COVID-19 is a type of disease that transmits a new variant of virus known as Severe Acute Respiratory Syndrome Coronavirus 2 (SARS-COV-2) in the same novel coronavirus family as SARS-CoV and Middle East Respiratory Syndrome Coronovirus (MERS-COV). A fast method to detect the disease is essential to prevent larger transmission and to look after the infected patients. The Chest X-ray, one of the detection methods of COVID-19 can be used in the examination process of suspected cases. In this paper, a COVID-19 detection model through chest x-ray images is proposed by using Grey Level Co-occurrence Matrix (GLCM) with Support Vector Machine (SVM), K-Nearest Neighbor (KNN), and Backpropagation Artificial Neural Network (BP-ANN) classifiers. In this case, Principal Component Analysis (PCA) will be added as a mean to optimize features extraction process. The aim of this work is to find the best classifier for predicting chest x-ray images as normal, pneumonia, or COVID-19 suspect. The BP-ANN emerged as the best classifier with 85,5% accuracy, 85,8% precision, and 86,1% recall. © 2023 Author(s).