Development of predictive models for personalized drug dosing in pediatrics DOI: https://doi.org/10.55522/jhpo.V2I2.0021

Abstract

 The safe and effective use of medications in pediatric populations presents unique challenges due to the physiological differences between children and adults. Standardized dosing regimens often fail to account for the variability in drug metabolism, distribution, and elimination in children. Personalized drug dosing, based on individual patient characteristics, is increasingly recognized as an essential strategy for improving therapeutic outcomes and minimizing adverse effects. The application of artificial intelligence (AI) and machine learning (ML) models has gained significant attention in recent years as potential solutions for developing predictive models for personalized drug dosing in pediatrics. By utilizing Big Data from electronic health records (EHRs), genetic data, and pharmacokinetic/pharmacodynamic (PK/PD) models, these models aim to optimize drug dosage based on a child’s unique clinical and genetic profile. This review explores the development, challenges, and current applications of predictive models for personalized pediatric drug dosing, highlighting the role of AI, ML, and Big Data in advancing this field. We also discuss the future potential of these technologies in enhancing pediatric pharmacotherapy and improving patient safety.