Skip to main content
Have a personal or library account? Click to login
Automated Detection and Severity Grading of Knee Osteoarthritis from X-ray Images Using Machine Learning with detailed literature Review Cover

Automated Detection and Severity Grading of Knee Osteoarthritis from X-ray Images Using Machine Learning with detailed literature Review

Open Access
|Sep 2026
Read online

Abstract

This paper presents a groundbreaking approach to the detection and severity grading of knee osteoarthritis (OA) utilizing X-ray imaging and a Convolutional Neural Network (CNN) architecture, specifically EfficientNet B5. Osteoarthritis, a prevalent degenerative joint disease, requires accurate and timely diagnosis for effective management. The proposedmethodology focuses on enhancing diagnostic accuracy through advanced deep learning techniques. The study leverages a diverse dataset of knee X-ray images sourced from various clinical settings, encompassing a spectrum of OA severity levels. The EfficientNet B5 architecture, known for its superior performance and efficiency, is employed for feature extraction and classification. The CNN model is trained and validated on a defined dataset, achieving an impressive accuracyof 95.84 percent in differentiating knee OA from normal conditions across multiple grades like Healthy, Minimal, Doubtful, Moderate, and Severe. The model further enhances clinical utility and incorporates a severity grading mechanism, providing clinicians with a detailed assessment of the disease progression. The severity grading is accomplished through fine-tuning the CNN model on a sub-dataset with annotated severity levels. The integration of severity grading demonstrates the model’s ability to detect knee OA accurately and quantify its severity with high precision. The findingsof this research highlight the potential of EfficientNet B5-based CNNs as a robust and efficient tool for knee OA diagnosis and severity grading. The reported accuracy of 95.84 percent positions the proposed model as a promising asset in clinical settings, offering a reliable and automated solution for early detection and precise evaluation of knee OA. This research, with its potential to significantly improve the diagnostic capabilities of musculoskeletal disorders, ultimately enhances patient care and outcomes by providing a more accurate and timely diagnosis.

DOI: https://doi.org/10.14313/jamris-2026-048 | Journal eISSN: 2080-2145 | Journal ISSN: 1897-8649
Language: English
Page range: 170 - 182
Submitted on: Aug 14, 2024
Accepted on: Apr 23, 2025
Published on: Sep 30, 2026
Published by: Łukasiewicz Research Network – Industrial Research Institute for Automation and Measurements PIAP
In partnership with: Paradigm Publishing Services
Publication frequency: 4 issues per year

© 2026 R. Gokulapriya, M. Balamurugan, Rakoth Kandan Sambandam, Divya Vetriveeran, published by Łukasiewicz Research Network – Industrial Research Institute for Automation and Measurements PIAP
This work is licensed under the Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 License.