AI-enabled ECG index for predicting left ventricular dysfunction in patients with ST-segment elevation myocardial...

Area of Science:

• Cardiology
• Artificial Intelligence
• Medical Diagnostics

Background:

• Electrocardiogram (ECG) changes post-primary percutaneous coronary intervention (PCI) in ST-segment elevation myocardial infarction (STEMI) correlate with patient prognosis.
• Predicting left ventricular (LV) dysfunction early is crucial for managing STEMI patients.

Purpose of the Study:

• To assess the feasibility of an AI-enabled ECG algorithm in predicting LV dysfunction in STEMI patients.
• To evaluate the AI algorithm's ability to quantify ECG changes after PCI and correlate them with clinical outcomes.

Main Methods:

• Serial ECGs from 637 STEMI patients were analyzed using an AI algorithm at multiple time points (pre-PCI, post-PCI, 6h, 24h, discharge, 1-month).
• The AI algorithm quantified STEMI probability and generated an index.
• Statistical analysis correlated the AI-derived probability index with the prevalence of LV dysfunction and clinical outcomes.

Main Results:

• The AI-derived probability index showed a significant association with the prevalence of LV dysfunction.
• A higher AI probability index independently predicted LV dysfunction.
• Elevated indices were linked to increased rates of cardiac death and heart failure hospitalizations.

Conclusions:

• AI-enabled ECG analysis effectively quantifies post-PCI ECG changes in STEMI patients.
• The AI-derived ECG index serves as a digital biomarker for predicting post-STEMI LV dysfunction, heart failure, and mortality.
• This AI tool aids in early identification of high-risk STEMI patients, facilitating timely interventions and improved outcomes.