Qiu, Haifeng
Liu, Li
Hong, Jiale
He, Yong
He, Lin
Luo, Yadong
Funding for this research was provided by:
Chongqing Social Science Planning Project (2023NDYB89)
Article History
Received: 8 October 2025
Accepted: 8 January 2026
First Online: 21 January 2026
Declarations
:
: The authors declare no competing interests.
: This study constitutes a retrospective analysis using exclusively publicly available, anonymized data; therefore, approval from an Institutional Review Board (IRB) was waived. The research utilized the GRAZPEDWRI-DX and FracAtlas datasets, both of which are distributed under the Creative Commons Attribution 4.0 International (CC BY 4.0) license. This license grants permission for the use, redistribution, and adaptation of the data for research purposes. All patient-identifiable information within these datasets was fully de-identified by the original data curators prior to public release, ensuring strict adherence to patient confidentiality and data privacy standards. The authors further declare that the deep learning model developed in this study is intended solely for scientific research and assistive diagnostic exploration. It has not been cleared for clinical use and must not be interpreted as a substitute for professional medical diagnosis, clinical judgment, or pathological confirmation.