Breaking binary in cardiovascular disease risk prediction.

Area of Science:

• Cardiovascular Medicine
• Genetics
• Artificial Intelligence

Background:

• Atherosclerotic cardiovascular disease (ASCVD) is a major global cause of mortality.
• Traditional risk assessment models have limitations in precision.
• Recent advancements in genetics, omics, and machine learning offer new avenues for risk stratification.

Purpose of the Study:

• To explore the role of polygenic risk scores (PRS) and machine learning (ML) algorithms in enhancing ASCVD risk stratification.
• To discuss the shift from binary risk prediction to continuum models.
• To highlight the need for addressing disparities in PRS implementation.

Main Methods:

• Integration of polygenic risk scores (PRS) derived from genomic data.
• Application of composite machine learning (ML) algorithms.
• Development of continuum-based risk models.

Main Results:

• Polygenic risk scores (PRS) and ML algorithms provide more comprehensive ASCVD risk stratification.
• These tools facilitate a move towards continuum models, offering nuanced risk assessment.
• The study identifies the necessity of addressing socioeconomic and racial disparities within PRS.

Conclusions:

• Polygenic risk scores (PRS) and machine learning (ML) represent a paradigm shift in ASCVD risk stratification.
• Continuum models derived from these technologies offer superior risk assessment compared to binary predictions.
• Equitable implementation of PRS is crucial to mitigate existing health disparities in cardiovascular disease.