Artificial Intelligence in Pediatric Orthopedics: A Comprehensive Review

Area of Science:

• Orthopedic Surgery
• Medical Imaging
• Artificial Intelligence

Background:

• Artificial intelligence (AI), including machine learning (ML) and deep learning (DL), is increasingly used in medicine.
• AI adoption in pediatric orthopedics is growing but lacks comprehensive review.
• This study reviews current AI applications in pediatric orthopedics.

Purpose of the Study:

• To systematically review the latest evidence on AI applications in pediatric orthopedics.
• To identify AI's role in diagnosing and managing pediatric orthopedic conditions.
• To highlight current limitations and future directions for AI in this field.

Main Methods:

• Literature search of PubMed and Web of Science databases up to March 2024.
• Selection of studies on AI in pediatric spinal deformities, hip disorders, trauma, bone age assessment, and limb discrepancies.
• Categorization of studies by AI application, models, datasets, and outcomes.

Main Results:

• AI models achieve high accuracy in spinal deformity classification (e.g., >90% with SVM, CNN).
• Deep learning excels in diagnosing hip dysplasia and detecting pediatric fractures (e.g., YOLO, ResNet).
• AI matches or surpasses traditional methods for bone age estimation, but studies often lack external validation and use small datasets.

Conclusions:

• AI has significant potential to improve diagnostic accuracy and decision-making in pediatric orthopedics.
• Current research faces limitations due to inconsistent methodologies and lack of standardized validation.
• Future research should prioritize multicenter data, prospective validation, and interdisciplinary collaboration for safe clinical integration.