Deep Learning Networks Accurately Detect ST-Segment Elevation Myocardial Infarction and Culprit Vessel

Area of Science:

• Cardiology
• Artificial Intelligence
• Medical Diagnostics

Background:

• Early diagnosis of ST-segment elevation myocardial infarction (STEMI) is crucial for better clinical outcomes.
• Identifying the specific blocked artery (culprit vessel) early also improves patient prognosis.

Purpose of the Study:

• To develop and evaluate deep learning (DL) models for detecting STEMI and its culprit vessels using 12-lead electrocardiography (ECG).
• To compare the diagnostic performance of DL models against experienced physicians.

Main Methods:

• Three DL models, including Convolutional Neural Network-Long Short-Term Memory (CNN-LSTM), were trained and tested on 883 cases (377 STEMI, 506 control).
• Model performance was assessed using Area Under the Curve (AUC) and compared with cardiologists, emergency physicians, and internists.

Main Results:

• The CNN-LSTM DL model achieved an AUC of 0.99 for STEMI detection, outperforming other DL models and physicians.
• DL models demonstrated high accuracy (AUC: 0.96) in identifying left anterior descending (LAD) artery occlusions, comparable to expert physicians.
• DL models were also comparable to physicians (AUC: 0.81) in distinguishing between right coronary artery (RCA) and left circumflex artery (LCX) occlusions.

Conclusions:

• ECG-based DL systems can effectively detect STEMI and predict culprit vessel occlusion.
• These AI-powered diagnostic systems enhance the accuracy and efficiency of STEMI diagnosis, potentially improving patient care.