Children and adolescents with chronic conditions, such as inflammatory bowel disease, juvenile idiopathic arthritis, or those undergoing long-term glucocorticoid therapy, are at a significantly increased risk of impaired bone health. This often manifests as low bone mineral density, elevated fracture risk, and accrual of peak bone mass. Traditionally, monitoring bone status in this vulnerable population has relied heavily on dual-energy X-ray absorptiometry (DXA), which, while the clinical gold standard, involves exposure to low-dose ionizing radiation. The need for repeated assessments to track disease progression or treatment response raises concerns regarding cumulative radiation exposure in the developing pediatric body. Furthermore, DXA provides primarily a two-dimensional (2D) areal density measurement and offers limited information on bone microarchitecture, a key determinant of bone strength. This clinical gap underscores the urgent need for safe, accessible, and more informative monitoring tools. The convergence of advanced, radiation-free imaging modalities—such as magnetic resonance imaging (MRI) and quantitative ultrasound—with sophisticated artificial intelligence (AI) for image analysis presents a transformative opportunity. This approach promises to enable frequent, detailed, and longitudinal assessment of bone quality and structure in pediatric patients without the associated radiation risks, potentially revolutionizing preventive care and management strategies.

Goal

The primary goal of this research initiative is to develop and implement a clinically viable, AI-assisted monitoring framework for bone health in pediatric patients with chronic conditions, thereby eliminating the need for additional ionizing radiation. This will be achieved by leveraging advanced, radiation-free imaging techniques, such as MRI or ultrasound, in combination with sophisticated artificial intelligence algorithms. The specific objectives are threefold: first, to train and validate AI models capable of automatically extracting quantitative biomarkers of bone density, microarchitecture, and strength from these medical images; second, to establish normative reference data and disease-specific trajectories for these AI-derived parameters in the pediatric population; and ultimately, to evaluate the clinical utility of this integrated system in enabling earlier detection of bone deterioration, improving risk stratification, and guiding timely therapeutic interventions, thereby improving long-term musculoskeletal outcomes for this at-risk group.

Scope and Information for Authors

This Research Topic seeks original research articles, reviews, and case studies focusing on innovative, non-ionizing approaches to pediatric bone health assessment. The scope explicitly encompasses the development and clinical application of AI algorithms for analyzing data from radiation-free modalities (e.g., MRI, ultrasound, HR-pQCT) in children with chronic diseases. Topics of interest include, but are not limited to:

● AI-based image segmentation and feature extraction for bone quality assessment;
● Validation of AI-derived biomarkers against clinical outcomes and existing gold standards;
● The establishment of pediatric reference databases for non-ionizing modalities;
● Studies on clinical integration, cost-effectiveness, and patient acceptance;
● Big data analysis of children's bone health

Interdisciplinary contributions bridging pediatrics, radiology, biomedical engineering, and computer science are highly encouraged. Authors should ensure their submissions clearly address the translational potential and specific relevance to the pediatric chronic disease population, adhering to the highest ethical standards for research involving minors.

Are you interested in contributing a manuscript to this Research Topic?

AI-Assisted Non-Ionizing Bone Health Assessment and Longitudinal Monitoring in Pediatric Chronic Diseases