Abstract
This study presents a comprehensive methodology that combines resampling and oversampling techniques to address the challenges of limited and balanced data, specifically in the context of viral emergencies such as the COVID-19 pandemic. Utilizing advanced statistical techniques like Bootstrap and SMOTE, the study conducts a retrospective analysis of COVID-19 patients, identifying those at higher risk of mortality. The proposed methodology not only enhances the accuracy of predictions in scenarios with limited data but also facilitates better decision-making in clinical triage systems. By applying these methods, the study achieves early and accurate identification of high-risk individuals, optimizing resource allocation and timely medical interventions. The results demonstrate that this combination of statistical techniques effectively improves health systems and responses to new viral threats, providing a robust foundation for informed decision-making in medical emergencies.

This publication has 25 references indexed in Scilit: