AI model successfully classifies knee prosthetic implants from plain radiographs with 100% accuracy: Study

Total knee arthroplasty (TKA) is considered the gold standard treatment for end-stage knee osteoarthritis. Common complications associated with TKA include implant loosening and periprosthetic fractures, which often require revision surgery or fixation. Challenges arise when medical records related to the knee prosthesis are lost, making it difficult to plan for revision surgery effectively. The study by --- et al developed an artificial intelligence (AI) system to classify the types of knee prosthetic implants using plain radiographs.

This retrospective experimental study included 7 types of knee prostheses commonly used in authors’ hospital. The artificial intelligence (AI) system was trained using YOLO (You Only Look Once) version 9, utilizing a dataset of 3228 post-operative and follow-up knee arthroplasty X-ray images. The plain radiographic images were augmented, resulting in a dataset of 25,800 images. Model parameters were fine-tuned to optimize performance for implant classification.

Key findings of the study were:

• The study included a total of seven types of knee prostheses:

NexGen (Zimmer Biomet), Persona (Zimmer Biomet), Vanguard

(Zimmer Biomet), Legion (Smith & Nephew), Gemini CR (Link), Gemini PS (Link), and Sigma PFC (Johnson & Johnson).

• The mean age of the patients was 62.8 years. Right knee arthroplasty was performed in 48.3% of cases, while left knee arthroplasty was performed in 51.7%.

• The images of knee prostheses comprised 50.9% of the dataset from the anteroposterior (AP) view and 49.1% from the lateral view.

• The AI model demonstrated exceptional performance metrics, achieving precision, recall, and accuracy rates of 100%, with an F1 score of 1.

• The area under the curve (AUC) of the receiver operating characteristic (ROC) curve was calculated to be 100%.

The authors concluded – “The AI model we developed achieves 100% accuracy in classifying types of knee prostheses. Its successful real-world clinical implementation aids surgeons in planning surgeries effectively. Future research should expand to include all types of knee implants worldwide, with the aim of creating a user-friendly application that is accessible globally to improve patient outcomes.”

Further reading:

AI classification of knee prostheses from plain radiographs and real‑world applications

Prin Twinprai et al

European Journal of Orthopaedic Surgery & Traumatology (2025) 35:107

https://doi.org/10.1007/s00590-025-04238-z