Related Experiment Video
Updated: Jul 29, 2026
Author Spotlight: AI-Driven Trypanosome Species Detection from Microscopic Images
Published on: October 27, 2023
Artificial intelligence and human expertise in hand trauma diagnosis: A collaborative approach
Céline Klein1, Pierre Fondu2, Daniel Aiham Ghazali3
1Department of Paediatric Orthopaedics, Jules Verne University of Picardie and Amiens Picardie University Hospital, Amiens, France; MP3CV MP3CV-EA7517, CURS, Amiens University Hospital and Jules Verne University of Picardie, Amiens, France.
Artificial intelligence (AI) shows high accuracy in diagnosing hand fractures and dislocations, comparable to experienced surgeons. While valuable for less experienced practitioners, AI requires further improvement for detecting dislocations and amputations.
Area of Science:
- Radiology
- Artificial Intelligence
- Orthopedic Surgery
Background:
- Hand injuries are common emergency department visits requiring radiographic analysis.
- Undiagnosed or misdiagnosed hand injuries can lead to poor functional outcomes.
- Artificial intelligence (AI) offers new diagnostic tools for clinical practice.
Purpose of the Study:
- To assess the diagnostic performance of AI in identifying hand fractures and dislocations compared to experienced hand surgeons.
- To evaluate the diagnostic performance of AI against that of a resident physician.
Main Methods:
- A retrospective study analyzed 1915 hand radiography datasets from patients over 16 years old.
- Two senior hand surgeons established the gold standard for diagnosis.
- AI and a resident physician's diagnoses were compared to the gold standard using sensitivity and specificity.
Main Results:
- The AI demonstrated high diagnostic accuracy with 97.6% sensitivity and 88.9% specificity.
- The AI's performance was comparable to senior surgeons but outperformed a resident physician in sensitivity.
- The AI failed to diagnose 11 injuries (0.6%), including dislocations and fractures.
Conclusions:
- AI can be a valuable tool in emergency settings, particularly for less experienced clinicians.
- AI does not surpass the diagnostic capabilities of senior hand surgeons.
- AI should complement, not replace, clinical examination, with improvements needed for dislocation and amputation detection.
