Machine Learning for E-triage

Authors : Şebnem Bora, Aylin Kantarci, Arife Erdoğan, Burak Beynek, Bita Kheibari, Vedat Evren, Mümin Alper Erdoğan, Fulya Kavak, Fatmanur Afyoncu, Cansu Eryaz, Hayriye Gönüllü

Pages : 86-90

View : 28 | Download : 9

Publication Date : 2022-07-20

Article Type : Research

Abstract :Due to the rising number of visits to emergency departments all around the world and the importance of emergency departments in hospitals, the accurate and timely evaluation of a patient in the emergency section is of great importance. In this regard, the correct triage of the emergency department also requires a high level of priority and sensitivity. Correct and timely triage of patients is vital to effective performance in the emergency department, and if the inappropriate level of triage is chosen, errors in patients' triage will have serious consequences. It can be difficult for medical staff to assess patients' priorities at times, therefore offering an intelligent method will be pivotal for both increasing the accuracy of patients' priorities and decreasing the waiting time for emergency patients. In this study, we evaluate the machine learning algorithms in triage procedure. Our experiments show that Random Forest approach outperforms the others in e-triage.
Keywords : Triage, Machine Learning, Emergency Department, Random Forest, Support Vector Machine, Decision Trees, Kth Nearest Neighbor