Artificial Intelligence in Cardiac MRI: Is Clinical Adoption Forthcoming?

Area of Science:

• Cardiovascular Magnetic Resonance (CMR) imaging
• Artificial Intelligence (AI) in medical diagnostics
• Deep learning applications in healthcare

Background:

• AI algorithms learn from training data to perform tasks requiring human intelligence, revolutionizing medical imaging.
• Cardiovascular Magnetic Resonance (CMR) is increasingly benefiting from AI for enhanced image acquisition, reconstruction, and analysis.
• Despite advancements, integrating AI into routine clinical practice for CMR remains a significant challenge.

Purpose of the Study:

• To review current AI principles and applications in CMR.
• To explore prospective observational studies utilizing AI in cardiology patient cohorts.
• To discuss the clinical utility of AI in CMR for diagnosis and prognostication.

Main Methods:

• Review of AI techniques applied to CMR data acquisition, reconstruction, and analysis.
• Analysis of clinical studies employing AI for undersampled reconstruction (cine, whole-heart, mapping, fingerprinting).
• Focus on AI-driven CMR prediction models for cardiac disease prognostication.

Main Results:

• AI offers deep learning solutions for expediting CMR scans and improving image analysis.
• AI-based biomarkers show promise for various cardiac conditions.
• Studies demonstrate AI's utility in post-processing and analysis, including prognostication models.

Conclusions:

• AI integration in CMR is rapidly advancing, with significant potential to support clinical decision-making.
• Further research is needed to validate AI tools with clinically relevant endpoints and large patient cohorts.
• Multi-disciplinary collaboration is crucial for advancing evidence-based medicine in AI-driven CMR.