Adv Clin Chem. 2026;135:227-251. doi: 10.1016/bs.acc.2026.08.004. Epub 2026 Aug 31.
ABSTRACT
Cardiovascular disease (CVD) is among the leading causes of morbidity and mortality worldwide. Advances in the development of novel diagnostic and therapeutic strategies are required to improve diagnostic accuracy and provide personalized treatments in CVD. Emerging research substantiates the role of genetic composition in CVD, prompting more comprehensive understanding via multimodal approaches that integrate clinical, genomic, and transcriptomic data. Traditional bioinformatics and statistical modelling have long been defaults for the analysis and interpretation of single-omic and clinical data. Implementation of reproducible, replicable, and transparent artificial intelligence (AI) and machine learning (ML) has the potential to revolutionize the investigation of high dimensionality and heterogeneity of multimodal data. As such, AI/ML can be helpful in discovering novel biomarkers and predicting disease susceptibility in general and CVD specifically. Herein, this review compares and discusses results produced using orthodox bioinformatics and statistics vs AI/ML methodology investigating whole genome, transcriptome and clinical data in CVD. Advances include emerging biomarkers associated with CVD (heart failure, atrial fibrillation, atherosclerosis, cardiomyopathy), others (multiple cancer disorders, anemia, type 1 diabetes mellitus, periodontal and rare diseases) and those common to both.
PMID:42843956 | DOI:10.1016/bs.acc.2026.08.004