Introduction: Artificial intelligence (AI) is transforming healthcare, and pediatric genetics has emerged as a particularly promising field for its implementation due to the complexity of rare genetic disorders and the rapidly increasing volume of genomic data. AI-based technologies may improve diagnostic accuracy, facilitate genomic interpretation, and support precision medicine approaches in pediatric patients.
Methods: This study was designed as a structured narrative review informed by a systematic literature search. A comprehensive search was conducted in PubMed, Scopus, and Web of Science databases for English-language publications published between January 2019 and January 2026. Relevant original studies, reviews, consensus statements, and ethically focused publications related to AI applications in pediatric genetics were included.
Results: AI-supported tools, including deep learning-based facial dysmorphology analysis and machine learning-driven genomic variant prioritization systems, have demonstrated significant potential in improving phenotype-genotype correlation, diagnostic efficiency, and personalized management strategies. AI applications also contribute to predictive modeling, disease-risk stratification, and large-scale multi-omics research. However, important ethical and legal concerns remain, including informed consent in minors, algorithmic bias, data privacy, transparency, and equitable access to AI technologies.
Discussion and Conclusion: AI has considerable potential to transform pediatric genetic practice by enhancing diagnostic precision and supporting individualized clinical care. Nevertheless, successful and equitable integration into routine practice requires robust ethical governance, prospective validation in diverse populations, transparent algorithms, and multidisciplinary clinician oversight. AI should be regarded as a decision-support tool that augments rather than replaces clinical expertise.
Keywords: Artificial intelligence; Dysmorphology; Ethical governance; Genomic medicine; Machine learning; Pediatric genetics; Rare diseases; Variant interpretation
