Diagnostic performance of artificial intelligence in traumatic femur fracture: A systematic review and meta-analysis.
SurgeryJul 3, 2026 (epub)
Audrey Bankes, Sergio M Navarro, Zhen Wang et al.
✦ AI-curated · Sources linked
In 30 seconds
This systematic review and meta-analysis evaluated the diagnostic accuracy of artificial intelligence (AI) in identifying acute traumatic femur fractures across 37 studies involving 95,148 femur images. The pooled sensitivity and specificity of AI were both 0.94, with an area under the receiver operating characteristic curve of 0.98, indicating high diagnostic performance comparable to human experts.
Key findings
- Pooled AI specificity was 0.94 (95% CI, 0.92-0.97) and sensitivity was 0.94 (95% CI, 0.92-0.96).
- AI achieved an area under the receiver operating characteristic curve of 0.98 (95% CI, 0.97-0.99).
- Expert readers showed an area under the curve of 0.989 unaided and 0.99 with AI assistance.
- Nonexpert readers improved from 0.95 unaided to 0.97 with AI augmentation.
Why it matters
The findings highlight the potential of artificial intelligence to enhance diagnostic accuracy in traumatic femur fractures, particularly for nonexpert clinicians. This could lead to improved patient outcomes through more accurate and timely diagnoses.