Artificial intelligence-enabled electrocardiography from scientific research to clinical application.

Area of Science:

• Cardiology
• Medical Diagnostics
• Artificial Intelligence

Background:

• Traditional electrocardiography (ECG) interpretation relies on established criteria.
• Human interpretation has limitations in detecting subtle patterns.
• Advancements in AI offer new diagnostic capabilities.

Purpose of the Study:

• To review the impact of AI on ECG analysis in cardiovascular diagnostics.
• To highlight how AI overcomes limitations of human interpretation.
• To discuss AI applications in risk stratification and screening.

Main Methods:

• Review of recent advancements in AI, particularly deep learning algorithms.
• Analysis of AI's capacity to process complex, high-dimensional ECG data.
• Examination of findings from randomized controlled trials (RCTs) on AI-ECG integration.

Main Results:

• AI models identify patterns missed by conventional methods, such as asymptomatic low ejection fraction and paroxysmal atrial fibrillation.
• AI integration in clinical workflows reduces intervention times.
• AI effectively identifies patients at elevated risk of adverse cardiovascular outcomes.

Conclusions:

• AI significantly enhances ECG's diagnostic utility for cardiovascular diseases.
• AI-ECG facilitates earlier clinical intervention and improved risk stratification.
• Future work includes integrating diverse data sources and enhancing AI model interpretability.