Proof-of-concept: Deep learning model estimates bone age from pediatric chest radiographs

Deep learning model estimated bone age from pediatric chest radiographs with intraclass correlation coefficient of 0.87 (95% CI 0.63-0.99) and root mean squared error of 1.30 years (95% CI 0.56-1.95) compared to hand radiograph reference.

• Retrospective proof-of-concept study with 237 chest radiographs from 101 children aged 3–15 years.
• Sex-considering deep learning model improved agreement (ICC 0.87 vs 0.81) and lowered error (1.30 vs 1.52 years) compared to sex-non-considering model.
• Findings suggest feasibility when dedicated hand radiographs are unavailable.

Automated summary

RadPigeon summaries are original and for information only. They are not clinical advice.