| Abstrakt: |
Aim: The COVID-19 pandemic, first originating in Wuhan, China in December 2019, has affected over 180 countries worldwide. The clinical spectrum of COVID-19 ranges from mild to severe pneumonia with acute respiratory distress syndrome. The sudden increase in COVID cases requiring hospitalization has made inpatient health institutions difficult to predict and manage. Machine learning models have been used to diagnose the disease, predict clinical course, and hospital stay. Materials and Methods: Data from 322 PCR-positive patients were analyzed, including demographics, comorbidities, laboratory values, and radiological results. Machine learning algorithms such as Logistic Regression, Support Vector Machine, Ensemble Methods, and K-Nearest Neighbor were used for classification. Results: Results showed that SVM provided the best classification performance. The model considered factors like age, gender, medical history, and test results to personalize treatment decisions. The study suggests that machine learning can improve patient care during the COVID-19 pandemic. Limitations include the need for validation with larger datasets from multiple centers. Conclusion: This study aimed to show whether machine learning techniques can be used to make decisions about the hospitalization of COVID-19 patients. [ABSTRACT FROM AUTHOR] |
