Early Detection of Acute Coronary Syndrome Using a Mobile Digital Health Application

Area of Science:

• Cardiology
• Artificial Intelligence in Medicine
• Digital Health

Background:

• Early detection of acute coronary syndrome (ACS) is critical for minimizing myocardial damage.
• Chest pain is the primary symptom necessitating rapid assessment.
• Mobile health applications offer novel avenues for timely diagnostic tools.

Purpose of the Study:

• To evaluate the efficacy of a mobile application-based questionnaire for predicting ACS.
• To assess the performance of a random forest machine learning model in identifying ACS cases.
• To determine the potential of the DETAK application for widespread early ACS detection.

Main Methods:

• A chest pain assessment questionnaire, developed by expert consensus, was integrated into the DETAK mobile application.
• Data from 566 patients (412 with ACS, 154 without) were analyzed.
• A random forest machine learning model (Python 3.12.4) was employed for ACS prediction.

Main Results:

• The machine learning model achieved high performance metrics: accuracy of 0.81, precision of 0.86, recall of 0.9, and an F1-score of 0.88.
• The model demonstrated a specificity of 0.54 for ACS prediction.
• The DETAK questionnaire provided a rapid and effective method for initial ACS assessment.

Conclusions:

• The DETAK mobile application's chest pain questionnaire, coupled with machine learning, shows significant potential for the early detection of acute coronary syndrome.
• This digital health tool can facilitate broader community-based screening and timely intervention for ACS.
• Further implementation of such tools could improve patient outcomes by reducing ischemic time.