Contributing Clinical Attributes to COVID-19 Mortality in Jakarta: Machine Learning Study

Aminanto, M. E.; Nasution, B. I.; Sulasikin, A.; Nugraha, Y.; Kanggrawan, J.; Suherman, A. L.

Since December 2019, we have lived in a pandemic era of severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2). Medical records of COVID-19 patients have been reported and analyzed worldwide. The Health Agency of Jakarta, Indonesia, collected clinical symptoms, demographics, travel history, and mortality information from March 2020 up to now. Despite massive research on COVID-19 patients' data, the significant clinical symptoms that lead to COVID-19 mortality in Jakarta have not been well described to the best of the authors' knowledge. We extracted the COVID-19 records in Jakarta and compared them between patients who were discharged and deceased. This paper examines each clinical symptom's importance to mortality using machine learning techniques, namely weighted Artificial Neural Network, Decision Tree, and Random Forest. We observed that Pneumonia, Shortness of Breath, Malaise, Hypertension, Fever, and Runny Nose are the top six significant clinical symptoms that lead to deaths in Jakarta. We suggest medical experts become more cautious with these symptoms. Also, in medical facilities, these symptoms can be used as prescreening before entering the facilities. © 2021 IEEE.