MRI AI system achieves high accuracy for olecranon bone age staging in adolescents

An automated MRI-based AI framework staged pubertal olecranon bone age with internal accuracies of 94.12% (girls) and 93.75% (boys); external validation at a second centre achieved 86.67% (girls) and 90.48% (boys), with mean absolute errors of 0.233 and 0.095 years.

• Cross-centre transportability evidence is preliminary; the tool is intended as a supplementary maturity marker, not a standalone forensic tool.
• Internal validation stage-mapped MAE/RMSE were 0.029/0.121 years (girls) and 0.062/0.250 years (boys).

Automated summary

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