Diagnostic accuracy of AI-assisted versus independent physician interpretation for bone fractures: a systematic review and meta-analysis.
Eur Radiol ExpSep 17, 2026 (epub)
Ming Zhang, Bihan Tang, Haoran Dai et al.
✦ AI-curated · Sources linked
In 30 seconds
This systematic review and meta-analysis evaluated the diagnostic performance of AI-assisted interpretation versus independent physician assessment for fracture detection. In 28 included studies, AI significantly improved pooled sensitivity to 87% compared to 73% for unassisted diagnosis, while maintaining high specificity. The study highlights the potential of AI to enhance diagnostic accuracy, particularly for junior clinicians.
Key findings
- AI-assisted diagnosis improved pooled sensitivity to 87% compared to 73% for independent assessment.
- Specificity remained high at 95% for AI-assisted interpretation versus 94% for unassisted.
- Junior clinicians experienced a 21% absolute increase in sensitivity with AI assistance.
- The SROC-AUC increased from 0.849 to 0.929 with AI support.
Why it matters
The integration of AI in fracture detection can significantly enhance diagnostic accuracy, particularly for less experienced clinicians, thereby reducing the risk of missed fractures and improving patient outcomes.