Artificial Intelligence in Pediatric Cardiovascular Genetics: From Multimodal Diagnosis to Risk Prediction and Precision Therapeutics
- 1Clinical Genetics Outpatient Clinic, Mother and Child Health Care Institute of Serbia “Dr Vukan Cupic”, 11070 Belgrade, Serbia
- 2Department of Cardiology, Mother and Child Health Care Institute of Serbia “Dr Vukan Cupic”, 11070 Belgrade, Serbia
- 3Department of Pediatrics, Faculty of Medicine, University of Belgrade, 11000 Belgrade, Serbia
Abstract
Artificial intelligence (AI) is increasingly being explored as a bridge between high-dimensional cardiovascular phenotyping and genomic interpretation in children with inherited and congenital heart disease. This review focuses specifically on where AI-assisted approaches in pediatric cardiovascular genetics have progressed beyond proof-of-concept, how the maturity of evidence differs between general genomic tools and pediatric cardiovascular applications, and what is required for responsible clinical translation. Inherited and rare cardiovascular disorders are characterized by genetic heterogeneity, incomplete penetrance, variable expressivity, and age-dependent phenotypes, while modern sequencing, imaging, and longitudinal monitoring generate data that are difficult to integrate using conventional approaches alone. AI-assisted methods are being investigated across diagnostic, prognostic, and therapeutic domains, including computational phenotyping, genomic variant prioritization, electrocardiographic and imaging analysis, multimodal risk prediction, and selection of candidates for precision therapies. The strongest pediatric cardiovascular evidence currently comes from externally evaluated ECG and echocardiographic models and from large outcome-prediction cohorts, whereas many genomic tools—including variant callers and pathogenicity predictors—remain general-purpose methods that have not been specifically validated in pediatric cardiovascular genetics. Recent disease-specific genomic models and automated reanalysis frameworks illustrate a shift toward more clinically contextualized interpretation, but current evidence remains limited by retrospective design, small or selectively assembled rare-disease cohorts, domain shift, population bias, incomplete calibration, and uncertain clinical utility. Accordingly, AI should be viewed as a governed decision-support layer rather than an autonomous authority. With external validation, transparent reporting, longitudinal monitoring, and multidisciplinary oversight, AI has the potential to augment specialist expertise and may support earlier diagnosis, improved risk stratification, and more individualized treatment.
Related
- Theranostic Imaging Agents: Integrating Molecular Diagnostics, Precision Therapeutics, and Artificial Intelligence
- Toward integrated multimodal artificial intelligence in bone marrow examination: current state and open challenges
- Artificial Intelligence–Driven Multimodal Imaging for Cancer During Pregnancy: Advances in Maternal–Fetal Diagnostics and Precision Oncology
- Artificial Intelligence, Multi-Omics, and Multimodal Integration in Precision Radiotherapy for Head and Neck Cancer: From Biological Insights to Treatment Decisions
- Artificial Intelligence for Cardiovascular Health in China
BibTeX
@article{ilic2026artificial,
title = {{Artificial Intelligence in Pediatric Cardiovascular Genetics: From Multimodal Diagnosis to Risk Prediction and Precision Therapeutics}},
author = {Nikola Ilić and Staša Krasić and Vladislav Vukomanović and Adrijan Sarajlija},
journal = {International Journal of Molecular Sciences},
year = {2026},
month = sep,
doi = {10.3390/ijms27188284},
url = {https://www.mdpi.com/1422-0067/27/18/8284},
note = {CC BY 4.0},
}
RIS
TY - JOUR
TI - Artificial Intelligence in Pediatric Cardiovascular Genetics: From Multimodal Diagnosis to Risk Prediction and Precision Therapeutics
AU - Nikola Ilić
AU - Staša Krasić
AU - Vladislav Vukomanović
AU - Adrijan Sarajlija
PY - 2026
DA - 2026/09/17
JO - International Journal of Molecular Sciences
DO - 10.3390/ijms27188284
UR - https://www.mdpi.com/1422-0067/27/18/8284
AB - Artificial intelligence (AI) is increasingly being explored as a bridge between high-dimensional cardiovascular phenotyping and genomic interpretation in children with inherited and congenital heart disease. This review focuses specifically on where AI-assisted approaches in pediatric cardiovascular genetics have progressed beyond proof-of-concept, how the maturity of evidence differs between general genomic tools and pediatric cardiovascular applications, and what is required for responsible clinical translation. Inherited and rare cardiovascular disorders are characterized by genetic heterogeneity, incomplete penetrance, variable expressivity, and age-dependent phenotypes, while modern sequencing, imaging, and longitudinal monitoring generate data that are difficult to integrate using conventional approaches alone. AI-assisted methods are being investigated across diagnostic, prognostic, and therapeutic domains, including computational phenotyping, genomic variant prioritization, electrocardiographic and imaging analysis, multimodal risk prediction, and selection of candidates for precision therapies. The strongest pediatric cardiovascular evidence currently comes from externally evaluated ECG and echocardiographic models and from large outcome-prediction cohorts, whereas many genomic tools—including variant callers and pathogenicity predictors—remain general-purpose methods that have not been specifically validated in pediatric cardiovascular genetics. Recent disease-specific genomic models and automated reanalysis frameworks illustrate a shift toward more clinically contextualized interpretation, but current evidence remains limited by retrospective design, small or selectively assembled rare-disease cohorts, domain shift, population bias, incomplete calibration, and uncertain clinical utility. Accordingly, AI should be viewed as a governed decision-support layer rather than an autonomous authority. With external validation, transparent reporting, longitudinal monitoring, and multidisciplinary oversight, AI has the potential to augment specialist expertise and may support earlier diagnosis, improved risk stratification, and more individualized treatment.
SN - 1422-0067
ER -