Advancing coronary artery disease risk stratification: integrating hyperspectral imaging with machine learning for...

Area of Science:

• Cardiology
• Medical Imaging
• Artificial Intelligence

Background:

• Coronary artery disease (CAD) is a major global cause of mortality.
• Accurate risk stratification is crucial for personalized cardiovascular medicine.
• Existing reviews on machine learning for CAD often overlook hyperspectral imaging (HSI).

Purpose of the Study:

• To exclusively focus on the synergistic potential of HSI and machine learning (ML) for advancing CAD diagnosis and risk stratification.
• To systematically review the current applications, technical advantages, and challenges of HSI in CAD diagnosis.
• To propose a novel noninvasive diagnostic framework integrating HSI and ML.

Main Methods:

• Systematic review of existing literature on HSI and ML in CAD.
• Analysis of HSI's technical advantages for noninvasive biomarker extraction.
• Integration of HSI data with ML algorithms for diagnostic framework development.

Main Results:

• HSI shows significant potential for noninvasive CAD detection and risk stratification.
• An integrated HSI-ML framework can extract CAD-related biomarkers from superficial body features.
• Key challenges include data standardization, model generalizability, and clinical translation.

Conclusions:

• The integration of HSI and ML presents a promising noninvasive approach for precision cardiology.
• This approach has the potential to significantly advance CAD diagnosis and risk stratification.
• Overcoming current challenges is essential for successful clinical translation.