Enhancing coronary artery plaque analysis via artificial intelligence-driven cardiovascular computed tomography

Area of Science:

• Cardiovascular Imaging
• Artificial Intelligence in Medicine
• Radiology

Background:

• Coronary computed tomography angiography (CCTA) is a key noninvasive tool for evaluating coronary artery disease (CAD).
• Conventional CCTA analysis has seen technological advancements, but limitations persist.
• Artificial intelligence (AI) integration promises to overcome these limitations.

Purpose of the Study:

• To evaluate the efficacy, accuracy, and efficiency of AI-driven CCTA (AI-CCTA) in CAD assessment.
• To highlight the advantages of AI-CCTA over conventional methods and invasive angiography.
• To discuss the current limitations and future potential of AI-CCTA.

Main Methods:

• Review of recent studies on AI-driven analysis of CCTA imaging.
• Comparison of AI-CCTA performance against conventional CCTA and invasive coronary angiography.
• Assessment of AI-CCTA capabilities in evaluating stenosis, plaque characteristics, and CT-derived fractional flow reserve.

Main Results:

• AI-CCTA significantly reduces radiation exposure to sub-millisievert levels.
• AI-CCTA demonstrates comparable accuracy and consistency to expert readers for coronary artery calcium scoring.
• AI-CCTA provides detailed plaque characterization and prognosticative value beyond luminal information.

Conclusions:

• AI-CCTA offers enhanced speed, consistency, accuracy, and safety in CAD evaluation.
• AI-CCTA presents a significant opportunity to advance cardiovascular care through data-driven insights.
• Further large-scale validation studies and AI model refinement are necessary to address current limitations.