Bridging Genomics to Cardiology Clinical Practice

Figure 1. The PRISMA Flow Chart of Study Screening and Selection AI = artificial intelligence; CVD = cardiovascular disease; PRISMA = Preferred Reporting Items for Systematic Reviews and Meta-Analyses; PRS = polygenic risk score.

Figure 2. Risk of Bias Assessment Using the QUADAS-2 Tool QUADAS-2 = Quality Assessment of Diagnostic Accuracy Studies-2.

Central Illustration. The Role of Artificial Intelligence in Optimizing Polygenic Risk Scores This figure highlights the role of ML algorithms in optimizing PRS for cardiovascular diseases. Data from 9 large biobanks were used to assess primary outcomes, including coronary artery disease (11 studies), myocardial infarction (1 study), and major adverse cardiovascular events (1 study). Key applications of AI tools in PRS optimization are presented: 1) improving feature selection with techniques such as Lasso regression and random forest; 2) integrating PRS with biomarkers, imaging, and clinical variables; and 3) combining multiple PRS using advanced methods like elastic net regression. The figure also compares the predictive performance of the best AI-optimized model with base PRS and clinical risk score models using AUC/C-index as comparison parameters. ASCVD PCE = atherosclerotic cardiovascular disease pooled cohort equation; AUC = area under the curve; CAD = coronary artery disease; MACE = major adverse cardiovascular event; MI = myocardial infarction; ML = machine learning; SNP = single nucleotide polymorphism; XGBoost = extreme gradient boosting; other abbreviations as in Figure 1.