Search

Introduction: Knee osteoarthritis is a prevalent condition frequently necessitating knee replacement surgery, with demand projected to rise substantially. Partial knee arthroplasty (PKA) offers advantages over total knee arthroplasty (TKA), yet its utilisation remains low despite guidance recommending consideration alongside TKA in shared decision making. Radiographic decision aids exist but are underutilised due to clinician time constraints., Materials and Methods: This research develops a novel radiographic artificial intelligence (AI) tool using a dataset of knee radiographs and a panel of expert orthopaedic surgeons' assessments. Six AI models were trained to identify PKA candidacy., Results: 1241 labelled four-view radiograph series were included. Models achieved statistically significant accuracies above random assignment, with EfficientNet-ES demonstrating the highest performance (AUC 95%, F1 score 83% and accuracy 80%)., Conclusions: The AI decision tool shows promise in identifying PKA candidates, potentially addressing underutilisation of this procedure. Its integration into clinical practice could enhance shared decision making and improve patient outcomes. Further validation and implementation studies are warranted to assess real-world utility and impact., Competing Interests: Declarations. Conflict of interest: The authors have no competing interests to declare that are relevant to the content of this article. Ethical approval: The study was approved by the UK National Research Ethics Service (London, UK, REC Reference: 18/CAG/0141), and performed in accordance with the ethical standards laid down by the 1964 Helsinki Declaration and its later amendments. Informed consent: This study only utilises retrospective anonymised patient images, and so informed consent was not deemed necessary by the UK National Research Ethics Service., (© 2024. The Author(s).)