Artificial Intelligence-Driven Integration of Pediatric Multi-Omics and Its Clinical Translation

Pediatric diseases exhibit unique pathophysiological characteristics, and traditional single-omics research struggles to capture their complex molecular mechanisms—impeding the advancement of accurate diagnosis. Pediatric multi-omics, which integrates genomics, transcriptomics, proteomics, and other omics layers, provides a holistic perspective on pediatric health, spanning common infections to rare genetic diseases. However, fragmented datasets, inconsistent standards, and challenges in cross-omics integration constrain its clinical translation, creating an urgent need for advanced tools such as artificial intelligence (AI).Leveraging machine learning (e.g., random forests) and deep learning (e.g., neural networks), AI excels at processing high-dimensional multi-omics data: it standardizes data formats, imputes missing values, and unravels hidden molecular correlations. Notably, it has improved the accuracy of pediatric brain tumor imaging by 15–20%, standardized multi-omics datasets for pediatric diabetes, and achieved 89% accuracy in differentiating asthma subtypes. Clinically, AI advances precision medicine—for instance, matching children with rare pediatric diseases to targeted therapies and increasing treatment response rates by 30%—and enables 48-hour-ahead prediction of neonatal sepsis, reducing mortality by 25%.Nevertheless, AI encounters key challenges: poor model interpretability (e.g., "black-box" issues) and insufficient pediatric-specific datasets. These can be addressed through interpretable AI approaches (e.g., SHAP values) and multi-center data consortia. Ethical concerns, particularly regarding genetic data privacy, also require mitigation through frameworks such as the FDA’s AI/ML Action Plan. Overall, AI-driven pediatric multi-omics is poised to transform pediatric healthcare; interdisciplinary collaboration to address these challenges will unlock its full potential, paving the way for precise and proactive pediatric care.To deepen understanding of AI research progress and applications in pediatrics, we invite submissions of Original Research articles, Reviews, Mini-Reviews, and special Case Reports. We are particularly interested in contributions focusing on: Application of AI tools for the diagnosis and differential diagnosis of pediatric diseases Utilization of AI tools for multi-omics analysis (encompassing genomics, proteomics, pathomics, radiomics, etc.) in clinical pediatric disease research Enhancement of prediction for childhood disease prevalence, diagnostic rates, and prognostic assessment using advanced AI algorithms to guide clinical treatment for pediatric children Segmentation of pediatric medical imaging data (e.g., ultrasound, computed tomography [CT], magnetic resonance imaging [MRI], and X-ray) via advanced AI algorithmsKeywords: Artificial Intelligence, Pediatric, Clinical Translation, Imaging, Integration Topic editors: Haiyan Yang, Yong Luo Source: https://www.frontiersin.org/research-topics/75247/artificial-intelligence-driven-integration-of-pediatric-multi-omics-and-its-clinical-translation