Skip to main page content
U.S. flag

An official website of the United States government

Dot gov

The .gov means it’s official.
Federal government websites often end in .gov or .mil. Before sharing sensitive information, make sure you’re on a federal government site.

Https

The site is secure.
The https:// ensures that you are connecting to the official website and that any information you provide is encrypted and transmitted securely.

Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
. 2026 Feb 18;108(4):303-312.
doi: 10.2106/JBJS.24.00579. Epub 2025 Nov 26.

Minimizing Missed Diagnoses of Tibial Plateau Fractures: The Role of AI in Radiographic Evaluation

Mu-Ze Chen  1 , Yueh-Peng Chen  2   3 , Yu-Chieh Hung  1 , Yu-Jui Huang  1 , Tzuo-Yau Fan  4 , Heng-Liang Liu  5 , Cheng-Pang Yang  1 , Shih-Sheng Chang  1 , Chang-Fu Kuo  2   3   6   7 , Cheng-Chi James Chu  2   4 , Yi-Sheng Chan  8
Affiliations

Minimizing Missed Diagnoses of Tibial Plateau Fractures: The Role of AI in Radiographic Evaluation

Mu-Ze Chen et al. J Bone Joint Surg Am. .

Abstract

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.

PubMed Disclaimer

Conflict of interest statement

Disclosure: Internal funding was received from Linkou Chang Gung Memorial Hospital (Chang Gung Medical Research Program, Grant No. CGRPG2M0021) and from the Center of Artificial Intelligence in Medicine, Chang Gung Memorial Hospital (Grant No. CLRPG3H0015). Disclosure of Potential Conflicts of Interest forms are provided with the online version of the article (http://links.lww.com/JBJS/I845).

Figures

Fig. 1
Fig. 1
Flow diagram of study patient selection. AP = anteroposterior, LAT = lateral, CGMH = Chang Gung Memorial Hospital, MSK = musculoskeletal.
Fig. 2
Fig. 2
The Heim system of squares.
Fig. 3
Fig. 3
Image-labeling example with the Heim system of squares.
Fig. 4
Fig. 4
Architecture for the deep learning multimodal classifier.
Fig. 5
Fig. 5
The AUC for detecting tibial plateau fractures in the test dataset.
Fig. 6
Fig. 6
The AUC for detecting tibial plateau fractures in the external validation dataset.
See this image and copyright information in PMC

References

    1. Rozell JC, Vemulapalli KC, Gary JL, Donegan DJ. Tibial Plateau Fractures in Elderly Patients. Geriatr Orthop Surg Rehabil. 2016. Sep;7(3):126-34. - PMC - PubMed
    1. Prasad N, Murray JM, Kumar D, Davies SG. Insufficiency fracture of the tibial plateau: an often missed diagnosis. Acta Orthop Belg. 2006. Oct;72(5):587-91. - PubMed
    1. LI S, et al. Diagnosis and Treatment of Tibial Plateau Fracture with Soft Tissue Injury. Chinese General Practice. 2020;23(11):1440.
    1. Guly HR. Diagnostic errors in an accident and emergency department. Emerg Med J. 2001. Jul;18(4):263-9. - PMC - PubMed
    1. Freed HA, Shields NN. Most frequently overlooked radiographically apparent fractures in a teaching hospital emergency department. Ann Emerg Med. 1984. Oct;13(10):900-4. - PubMed

LinkOut - more resources