Convolutional neural networks in paediatric fracture detection: pooled evidence from a systematic review and meta-analysis

Share

  • 1

    Fractures are common in children, with significant morbidity, making accurate and timely diagnosis essential to prevent complications.

  • 2

    AI, particularly convolutional neural networks, shows promise in improving diagnostic accuracy for paediatric fractures on X-rays.

  • 3

    Current research on AI for paediatric fracture detection is limited, with concerns about heterogeneity and generalizability of findings.

  • 4

    The systematic review aimed to evaluate the diagnostic accuracy of AI models for detecting paediatric appendicular fractures.

  • 5

    The study followed PRISMA-DTA guidelines and focused on sensitivity, specificity, and other accuracy metrics per radiograph.

Original Source(s)

Convolutional neural networks in paediatric fracture detection: pooled evidence from a systematic review and meta-analysis

  • by Alina Pervez, S. Umar Hasan, Alan R. Norrish

  • March 21, 2026

Related Content

Hypofractionated Locoregional RT in Early Breast Cancer

As reported in The Lancet by Rivera et al, 5-year results of a French phase III trial (UNICANCER HypoG-01) showed noninferiority of hypofractionated locoregional radiotherapy (RT) vs standard 5-week R...

Trauma CT and Unexpected Findings

Investigators evaluate classification of incidental findings on trauma whole-body CT.

Can AI Clarify Lung Screening?

Study assesses comprehension of large language model-generated lung screening summaries