Journal of Bone and Joint Surgery, American Volume最新文献

Background: Tibial plateau fractures represent a diverse group of intra-articular injuries that can be difficult to detect and characterize on initial imaging. The aim of the present study was to develop an artificial intelligence (AI) diagnostic tool for identifying tibial plateau fractures on radiographs.

Methods: In this retrospective study, we analyzed radiographs that had been made from January 2018 to December 2020 for 1,809 patients, with an equal distribution of male and female adults. A total of 3,821 anteroposterior and lateral knee radiographs were evaluated with use of the EfficientNet B3 AI model, with computed tomography (CT) images being used as the ground truth. Evaluation metrics focused on the area under the receiver operating characteristic curve (AUC) and positive predictive values across different subgroups.

Results: Our AI model attained AUCs of 0.98 and 0.97 for detecting tibial plateau fractures in the test and external validation datasets, respectively. Subgroup analysis revealed diverse positive predictive values across different Schatzker types and 3-column classifications.

Conclusions: Our deep learning model exhibits newfound ability for identifying tibial plateau fractures. However, we encountered several limitations, such as imbalances among the sizes of various subgroups in the dataset and an inability to identify radiographs containing foreign objects or other defects.

Level of evidence: Diagnostic Level III. See Instructions for Authors for a complete description of levels of evidence.

Background: The integration of artificial intelligence (AI) in orthopaedics and sports medicine (OSM) has transformed clinical practice and scientific inquiry. However, the increasing reliance on AI raises critical concerns regarding transparency, ethical considerations, and reproducibility. The aim of this study was to systematically evaluate the editorial policies of leading OSM journals concerning AI usage and the endorsement of AI-specific reporting guidelines (RGs).

Methods: A cross-sectional review was conducted in accordance with STROBE (Strengthening the Reporting of Observational Studies in Epidemiology) guidelines. The top 100 peer-reviewed OSM journals were identified using the 2023 SCImago Journal Rank (SJR). Data extraction included journal characteristics, AI-related policies within Instructions for Authors, and references to AI-specific RGs. Data were collected in a masked, duplicate fashion, with discrepancies resolved through consensus.

Results: Of the 100 journals analyzed, 94% referenced AI in their editorial policies, all of which explicitly prohibited AI authorship and required the disclosure of AI use in manuscript preparation. AI-generated content was permitted in 82% of journals. AI-assisted image generation was permitted by 60% of journals and explicitly prohibited by 34%. Despite these policies, only 1% of journals referenced AI-specific RGs, with the Checklist for Artificial Intelligence in Medical Imaging (CLAIM) being the sole guideline mentioned.

Conclusions: While most of the OSM journals had established policies on AI usage, there was a notable lack of standardization, particularly with respect to AI-generated images. Additionally, the absence of AI-specific RG endorsements highlights a gap in methodological guidance. Standardizing AI policies and encouraging the adoption of RGs could enhance the transparency, reproducibility, and ethical integrity of AI-driven research in OSM.