Converging artificial intelligence and multi-omics to elucidate the molecular landscape of congenital heart disease: A narrative review

Congenital heart disease (CHD), a major cause of fetal and neonatal morbidity and mortality, is the most prevalent birth defect worldwide. Despite the extensive progress achieved in genetic engineering and high-resolution imaging methods, more than 60% of cases of CHD remain unascertained. Artificial intelligence (AI) is rapidly emerging as an efficient instrument in CHD research and management, as it continues to advance risk analysis, enhance image processing, and facilitate personalized prognosis modeling. Today, AI is actively implemented to combine various forms of omics data and generate holistic information concerning complex datasets, including the results obtained through genomic, transcriptomic, proteomic, metabolomic, and epigenomic research, to better understand the multifaceted etiology of CHD. The purpose of this narrative review is to synthesize evidence from the past decade in the areas of developmental biology, genomics, and clinical cardiology and illustrate how the convergence of AI and multi-omics data could redefine the study of CHD. Specifically, such convergence could accelerate the discovery of molecular mechanisms and address health disparities by providing scalable, low-cost, accurate diagnostic tools for use in resource-poor environments. Additionally, the interplay between genetic, environmental, and epigenetic factors in the development of CHD is discussed, along with the current AI-based tools being used to transform diagnostic and research models in CHD. Lastly, a summary of AI-discovered genes and biomarkers is provided, and the possibility of computational intelligence discovering novel therapeutic targets and enabling the development of precise cardiology is suggested.