Quantification of mitral regurgitation: from traditional methods to artificial intelligence

Area of Science:

• Cardiology
• Medical Imaging
• Artificial Intelligence

Background:

• Mitral regurgitation (MR) is a common heart valve disorder linked to poor outcomes.
• Standard echocardiography methods for MR assessment have limitations, especially with complex or eccentric regurgitant jets.
• Accurate MR quantification is crucial for timely diagnosis and treatment.

Purpose of the Study:

• To evaluate the capabilities of artificial intelligence (AI) in quantifying mitral regurgitation (MR).
• To compare AI-based MR quantification with traditional echocardiographic methods and cardiac magnetic resonance.
• To explore the potential of AI for MR screening in underserved areas.

Main Methods:

• Utilized machine learning algorithms for automated analysis of echocardiographic data.
• Developed AI models to analyze regurgitant flow throughout systole, considering non-hemispheric orifice area and jet morphology.
• Validated AI quantification against established cardiac magnetic resonance imaging standards.

Main Results:

• AI-based MR quantification demonstrated favorable agreement with cardiac magnetic resonance.
• AI methods overcome limitations of traditional echocardiography, particularly for complex MR jet types.
• AI shows potential for direct MR detection and grading from echocardiographic clips.

Conclusions:

• AI represents a significant advancement in the accurate quantification of mitral regurgitation.
• Automated MR assessment using AI can improve diagnostic accuracy and potentially expand access to care.
• AI holds promise for MR screening, especially in settings with limited access to specialist expertise.