Objectives/Goals: The primary objective of this research is to examine current implementations of Artificial Intelligence (AI) models in clinical trials, focusing on advantages, limitations, and ethical considerations associated with their application in pediatric populations. Methods/Study Population: A comprehensive literature search was conducted using PubMed and Google Scholar databases. The initial search used the key terms “artificial intelligence” and “pediatric.” The results were refined to include publications from 2023 to 2025. The keyword “imaging” was added, focusing on studies exploring AI-based diagnostic tools. Articles were included if they had at least one of the following: (1) the need for pediatric-specific databases, (2) diagnostic accuracy within pediatric populations, or (3) ethical considerations in AI-driven pediatric research. Data were retrieved from ClinicalTrials.gov, where “artificial intelligence” was specified. Eligibility criteria were limited to pediatric participants and restricted to studies conducted in the USA to ensure regulatory consistency. Results/Anticipated Results: An analysis of trials from ClinicalTrials.gov further highlighted the underrepresentation of pediatric subjects in AI-related clinical research, emphasizing a critical gap in age-appropriate AI model development and validation. AI models demonstrate great potential for improving diagnostic accuracy, but effectiveness remains limited by lack of pediatric-specific training data. In pediatric chest imaging, AI algorithms achieved diagnostic performance comparable to adult populations in detecting bacterial pneumonia (Morcos et al., 2023). Conversely, a large-scale review of public medical imaging datasets revealed an increase in false-positive rates among younger patients, reflecting model performance disparities (Zamora et al., 2024). Discussion/Significance of Impact: Implementing AI-driven models can accelerate data analysis and streamline trial design. The application of AI models in pediatric clinical trials can accelerate the timeline for treatment and reduce misdiagnosis. Addressing data gaps and ethical considerations about consent and algorithmic bias are crucial to improve pediatric health outcomes.
Tsogmagnai et al. (Wed,) studied this question.