Emerging Artificial Intelligence Tools for the Screening of Structural and Valvular Heart Disease

Area of Science:

• Cardiology
• Medical Imaging
• Artificial Intelligence

Background:

• Structural heart disease (SHD) and valvular heart disease (VHD) affect heart valves, chambers, walls, and muscles.
• Current diagnostic methods for SHD and VHD have limitations in accessibility and predictive accuracy.

Purpose of the Study:

• To review recent advancements in AI-guided tools for screening SHD and VHD.
• To discuss the challenges and opportunities of implementing these AI tools in clinical practice.

Main Methods:

• Review of AI-guided models trained on electrocardiograms (ECGs), chest X-rays, and coronary artery calcium scans.
• Analysis of AI model accuracy in diagnosing SHD, heart failure, low ejection fraction, and VHD.

Main Results:

• AI-guided models demonstrate high accuracy in diagnosing various cardiac conditions, including SHD and VHD.
• Some AI models offer improved explainability by highlighting predictive signals.
• AI screening has the potential for earlier diagnosis, enhanced accuracy, and improved accessibility for SHD and VHD.

Conclusions:

• AI tools show promise for earlier and more accurate screening of structural and valvular heart disease.
• Further clinical outcome data is necessary to support broad deployment of AI in SHD and VHD screening.