A deep learning algorithm to detect proximal humerus fractures on radiographs
The DL algorithm developed in this study displayed robust and trustworthy performance when detecting the presence of a proximal humerus fracture on radiographs. This algorithm could be implemented in busy emergency department practices so that proximal humerus radiographs with a high probability of fracture can be flagged for dedicated review. In addition, this algorithm provides the first step to further develop other artificial intelligence tools to better understand and manage proximal humerus fractures.
Authors
- Linjun Yang (ORCID: https://orcid.org/0000-0003-1430-3355)
- Rodrigo de Marinis (ORCID: https://orcid.org/0000-0003-1888-8928)
- Sami Saniei (ORCID: https://orcid.org/0000-0002-7218-5695)
- John W. Sperling (ORCID: https://orcid.org/0000-0002-6569-7860)
- Joaquín Sánchez‐Sotelo (ORCID: https://orcid.org/0000-0003-3199-3247)
- Miguel M Girod (ORCID: https://orcid.org/0000-0002-2114-6303)
- Monty Khela (ORCID: https://orcid.org/0009-0004-4687-8588)
- Elizabeth S Kaji (ORCID: https://orcid.org/0009-0009-8745-7612)
Institutions
- Pontificia Universidad Católica de Chile (CL)
- Mayo Clinic in Arizona (US)
- Complejo Asistencial Sótero del Río (CL)
- Mayo Clinic in Florida (US)
Publication Details
- Journal
- JSES Reviews Reports and Techniques
- Published
- 2025-08-18
- DOI
- https://doi.org/10.1016/j.xrrt.2025.07.025
- Primary Topic
- Artificial Intelligence in Healthcare and Education
- Type
- article
- Field-Weighted Citation Impact
- 0.00