New research related to artificial intelligence (AI) is presented in the latest JBJS Specialty Update What’s New in Machine Learning and Generative Artificial Intelligence in Orthopaedics. Here, we summarize the 5 most impactful publications, as selected by coauthor Jason Strelzow, MD, FRCSC.
Risks and Prognostic Modeling in Orthopaedics
A convolutional neural network (CNN) was created to predict operative vs. nonoperative treatment of an isolated distal radial fracture in patients <60 years old on the basis of radiographs. A total of 163 patients were included. The model’s prediction accuracy was 88%1. “This highly discriminative triage prediction tool hints at the potential power of AI tools to augment existing care pathways, treatment algorithms, and triage protocols,” write Specialty Update authors Drs. Strelzow and Ghert.
The use of machine learning (ML) to predict osteoarthritis progression was evaluated in a systematic review of 39 studies. Predicting OA progression with use of ML models was shown to be feasible, but the clinical applicability of those models was limited by reused data sets and nonstandardized definitions of outcomes. The authors of the review emphasized the need for diverse data sources, standardized definitions and metrics, rigorous validation, and more sophisticated algorithms in future research2.
Editorial and Academic Writing
Guidelines for AI use in orthopaedic scholarly publications were established in a 2023 joint editorial by The Journal of Bone & Joint Surgery, Clinical Orthopaedics and Related Research, The Bone & Joint Journal, and the Journal of Orthopaedic Research. The journals determined that AI tools cannot be listed as authors, that AI use should be disclosed, and that only human authors can be held accountable for the accuracy of a work. The editors-in-chief acknowledged the benefits of AI use as well as the dangers, warning that “misuse of these tools can undermine the integrity of the scholarly record.”3
Data Quality, Bias, and the “Human Element”
Recently, scholars have been emphasizing the importance of external validation and bias assessment in ML studies, including the best practices for reporting orthopaedic research that uses AI4. The quality of the data input has also been stressed: in a review of deep learning (DL) in orthopaedics, researchers noted that model accuracy, explainability, and fairness are essential to ensuring that the model’s outputs are trustworthy5. Drs. Strelzow and Ghert explain, “Many risk factors in orthopaedics are measured inconsistently or contain implicit biases: for example, classification grades that vary between examiners or patient-reported outcome scores that can be influenced by demographic factors. If such noisy or biased data are used to train an algorithm, the predictions will likewise be unreliable or biased.”
What’s New in Machine Learning and Generative Artificial Intelligence in Orthopaedics is freely available at JBJS.org.
What’s New by Subspecialty
Each month, JBJS publishes a review of the most pertinent studies from the orthopaedic literature in a select subspecialty. To read the reports, visit the What’s New by Subspecialty collection at JBJS.org.
Recent OrthoBuzz posts include: What’s New in Pediatric Orthopaedics, What’s New in Adult Reconstructive Knee Surgery, and What’s New in Musculoskeletal Basic Science.
References
- Hsu D, Persitz J, Noori A, Zhang H, Mashouri P, Shah R, Chan A, Madani A, Paul R. Predicting surgical versus nonsurgical management of acute isolated distal radius fractures in patients under age 60 using a convolutional neural network. J Hand Surg Am. 2025 Jul;50(7):781-9.
- Castagno S, Gompels B, Strangmark E, Robertson-Waters E, Birch M, van der Schaar M, McCaskie AW. Understanding the role of machine learning in predicting progression of osteoarthritis. Bone Joint J. 2024 Nov 1;106-B(11):1216-22.
- Leopold SS, Haddad FS, Sandell LJ, Swiontkowski M. Editorial: Artificial intelligence applications and scholarly publication in orthopaedic surgery. Clin Orthop Relat Res. 2023 Jun 1;481(6):1055-6.
- Wyles CC, Saniei S, Mulford KL, Girod MM, Taunton MJ. Reporting guidelines for artificial intelligence use in orthopaedic surgery research. J Arthroplasty. 2025 Oct;40(10):2737-2743.e1
- Alzubaidi L, Al-Dulaimi K, Salhi A, Alammar Z, Fadhel MA, Albahri AS, Alamoodi AH, Albahri OS, Hasan AF, Bai J, Gilliland L, Peng J, Branni M, Shuker T, Cutbush K, Santamaría J, Moreira C, Ouyang C, Duan Y, Manoufali M, Jomaa M, Gupta A, Abbosh A, Gu Y. Comprehensive review of deep learning in orthopaedics: applications, challenges, trustworthiness, and fusion. Artif Intell Med. 2024 Sep;155:102935.
