Purpose: To explore the potential of ChatGPT-4o in analysing radiographic images of knee osteoarthritis (OA) and to assess its grading accuracy, feature identification and reliability, thereby helping surgeons to improve diagnostic accuracy and efficiency.
Methods: A total of 117 anterior‒posterior knee radiographs from patients (23.1% men, 76.9% women, mean age 69.7 ± 7.99 years) were analysed. Two senior orthopaedic surgeons and ChatGPT-4o independently graded images with the Kellgren-Lawrence (K-L), Ahlbäck and International Knee Documentation Committee (IKDC) systems. A consensus reference standard was established by a third radiologist. ChatGPT-4o's performance metrics (accuracy, precision, recall and F1 score) were calculated, and its reliability was assessed via two evaluations separated by a 2-week interval, with intraclass correlation coefficients (ICCs) determined.
Results: ChatGPT-4o achieved a 100% identification rate for knee radiographs and demonstrated strong binary classification performance (precision: 0.95, recall: 0.83, F score: 0.88). However, its detailed grading accuracy (35%) was substantially lower than that of surgeons (89.6%). Severe underestimation of OA severity occurred in 49.3% of the cases. Interrater reliability for surgeons was excellent (ICC: 0.78-0.91), whereas ChatGPT-4o showed poor initial consistency (ICC: 0.16-0.28), improving marginally in the second evaluation (ICC: 0.22-0.39).
Conclusion: ChatGPT-4o has the potential to rapidly identify and binary classify knee OA on radiographs. However, its detailed grading accuracy remains suboptimal, with a notable tendency to underestimate severe cases. This limits its current clinical utility for precise staging. Future research should focus on optimising its grading performance and improving accuracy to enhance diagnostic reliability.
Level Of Evidence: Level III retrospective comparative study.
Download full-text PDF |
Source |
|---|---|
| http://dx.doi.org/10.1002/ksa.12639 | DOI Listing |
Publication Analysis
Top Keywords
Similar Publications
Deep learning model for the early prediction of pathologic response following neoadjuvant chemotherapy in breast cancer patients using dynamic contrast-enhanced MRI.
Front Oncol
February 2025
Department of Radiation Oncology, The Affiliated Hospital of Qingdao University, Qingdao, Shandong, China.
Purpose: This study aims to investigate the diagnostic accuracy of various deep learning methods on DCE-MRI, in order to provide a simple and accessible tool for predicting pathologic response of NAC in breast cancer patients.
Methods: In this study, we enrolled 313 breast cancer patients who had complete DCE-MRI data and underwent NAC followed by breast surgery. According to Miller-Payne criteria, the efficacy of NAC was categorized into two groups: the patients achieved grade 1-3 of Miller-Payne criteria were classified as the non-responders, while patients achieved grade 4-5 of Miller-Payne criteria were classified as responders.
Nomograms of risk prediction and prognosis for the T1-T2 stage gastric cancer with lymph node metastasis: a population-based study.
Front Med (Lausanne)
February 2025
Department of Colorectal Hernia Surgery, Binzhou Medical University Hospital, Binzhou, China.
Background And Aims: Lymph node metastasis plays a crucial role in determining the appropriate treatment approach for patients with gastric cancer (GC), particularly those in the T1-T2 stage. Currently available diagnostic strategies for GC with lymph nodes have limited accuracy. The present research aimed to create and validate diagnostic and prognostic nomograms specifically tailored for the T1-T2 stage GC patients with LNM.
View Article and Find Full Text PDFDataset of 37-subject EEG recordings using a low-cost mobile EEG headset during a semantic relatedness judgment task.
Data Brief
April 2025
Psychology Department, Middle Tennessee State University, Murfreesboro, TN, USA.
This data article presents electroencephalography (EEG) data and behavioral responses from a study examining the efficacy of a consumer-grade EEG headset (InteraXon Muse 2) in measuring the N400 component, a neural marker of semantic processing. These data are linked to the article "Exploring the Utility of the Muse Headset for Capturing the N400: Dependability and Single-Trial Analysis". Data were collected from 37 adult native speakers of English while they completed a semantic relatedness judgment task.
View Article and Find Full Text PDFA Nomogram Based on S100A7 and Clinicopathological Characteristics to Predict the Efficacy of Neoadjuvant Chemotherapy in Breast Cancer: A Retrospective Study.
Ther Clin Risk Manag
March 2025
Department of Breast Disease Center, The Fourth Hospital of Hebei Medical University, Shijiazhuang, People's Republic of China.
Background: We have previously found that S100 calcium-binding protein A7 (S100A7) is strongly associated with chemoresistance in breast cancer (BC). In this study, we investigated whether S100A7 can be used to predict the efficacy of neoadjuvant chemotherapy (NAC) and assessed its relationship with clinicopathological characteristics in BC.
Methods: We retrospectively analyzed the clinicopathological data of patients with BC who underwent NAC at the Fourth Hospital of Hebei Medical University between January 2021 and December 2021.
The Value of P16 Protein Detection in the Screening for High-Grade Squamous Intraepithelial and Higher Lesions of the Cervix by the Combined Detection of HR-HPV and TCT.
Diagn Cytopathol
March 2025
Department of Pathology, Hunan Maternal and Child Health Hospital, Changsha, China.
Liquid-based thin layer cytology (TCT) and HR-HPV detection are the most important screening methods for cervical cancer. These two methods have limited sensitivity and specificity, so some cervical lesions are still missed or misdiagnosed. This paper mainly discusses the value of P16 protein detection in cervical cancer screening.
View Article and Find Full Text PDFWant AI Summaries of new PubMed Abstracts delivered to your In-box?
Enter search terms and have AI summaries delivered each week - change queries or unsubscribe any time!