Abstract
COVID-19 causes a wide spectrum of symptoms, such as mild upper respiratory infection or life-threatening sepsis. From 20.2% of cases of COVID-19 progressed to severe disease with a mortality rate of 3.1% where 60%-90% of patients with comorbidities were hospitalized. The purpose of this study was to find out that cluster analysis using K-Medoids can distinguish COVID-19 patients at various age levels which analytical method has sensitivity and specificity values in analyzing clustering in COVID-19 patients. This study uses a cohort retrospective design conducted at five hospitals in Yogyakarta Province. The study used patient medical record data from March 2020 – September 2021 with a total of 916 patient data that met the inclusion criteria. Cluster analysis will be carried out using Google Colaboratory with the Python programming language. The clustering results are divided into 2 cluster groups where cluster 1 consists of 558 patients and cluster 2 consists of 358 patients with various age levels. The test resulted in 2 clusters with a DBI value of 5,191631. The results of statistical tests showed that there was a significant relationship (p-value = 0,023) between age, recovery rate, and patient mortality. From the test results, it can be seen that ages 50 to 59 years are suspected of COVID-19
Published Version (
Free)
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call