Mifetika Lukitasari, Allen Lamarca Nazareno, Mohammad Saifur Rohman, Jitendra Jonnagaddala
Early detection of acute coronary syndrome (ACS) is vital for reducing ischemic time and preserving more heart muscle.Chest pain is the most common symptom of acute coronary syndrome (ACS). This study used a quick chest pain assessment questionnaire embedded in the DETAK mobile application to predict ACS. Data from 566 patients (412 with ACS and 154 without ACS) were analysed. Cardiologists confirmed the diagnosis of acute coronary syndrome (STEMI and NSTEMI). Patients completed the questionnaire, developed by expert consensus, within 48 hours of admission or transfer. Random forest machine learning, using Python version 3.12.4, was utilized to predict ACS. The model achieved an accuracy of 0.81, precision of 0.86, recall of 0.9, specificity of 0.54, and an F1-score of 0.88. This simple and quick assessment using DETAK shows the potential for scaling up the early detection of ACS in a broader community. © 2025 The Authors.
School of Population Health, UNSW Sydney, 2052, NSW, Australia; Institute of Mathematical Sciences, College of Arts and Sciences, University of the Philippines, Laguna, Los Banos, Philippines; Department of Cardiology and Vascular Medicine, Faculty of Medicine, Brawijaya University, Malang, Indonesia; Cardiovascular Research Centre, Brawijaya University, Malang, Indonesia
