Emergency triage classification with machine learning: A Colombian case

This study presents innovative research on classifying emergency department patients using advanced machine-learning techniques. Objective. Provide a decision-support tool for identifying patients who require urgent intervention promptly. Method. A balanced Random Forest model was developed, which showed promising triage results. The approach created triage subcategories and assigned different priorities based on critical status. Results. The results were encouraging, with an accuracy of 80.14% for high-priority patients, 79.45% for referred patients, and 81.29% for patients who ultimately died. Conclusion. The findings support the model's effectiveness in improving decision-making in emergency departments. The research aims to enhance patient classification efficiency, optimize resource allocation, and ensure timely care. It provides new insights and can benefit healthcare professionals by improving the quality of emergency care.