ABSTRACT
Artificial Intelligence (AI) has made significant advances in all aspects of healthcare. Different studies have been investigated using AI regarding improve patient diagnosis, emergency care, and patient safety for different diseases. The coronavirus disease (COVID-19) has escalated into a global public health emergency. Different have shown clinical decisions for COVID-19 on diagnosis accuracy, severity risk, and so on. Moreover, few studies have focussed on clinical decisions for COVID-19. The fact global pandemic of corona poses a big challenge for clinicians, this research intends to detect a coronavirus patient path based on the virus' biological traits and presents an adequate mechanism for the efficient decision support system that assists doctors in predicting a COVID-19 patient. To train and test our model, we used 311 patient data which are 214 (69%) male and 96 (31%) female. Data were collected and maintained from three centers of Ethiopian Hospitals from Nov 2021 to March 2022 with the age range of 21 up to 67. The experiments were performed with three Machine Learning (ML) algorithms, namely Naïve Bayes (NB), Artificial Neural Network (ANN), and Support Vector Machine (SVM). The model was validated using input variables (n=10) which achieved better involvement in the virus symptoms identification and based on the evaluation metrics, ANN achieved 97%, 96%, 85%, and 98.3% for recall, precision, and F1- measures respectively. Our results demonstrated that the SVM algorithm achieved a 91.3% average accuracy and the other two methods had 87.75% and 96.05% respectively. In this research, ANN does better than the NB classifier by 8.3% on average and better than SVM by 4.75%. In addition, using more prediction algorithms and a larger dataset includes more parameters used to estimate patients. [ FROM AUTHOR] Copyright of Journal of Pharmaceutical Negative Results is the property of ResearchTrentz and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use. This may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full . (Copyright applies to all s.)