IRO Journals
Journal of Artificial Intelligence and Capsule Networks

YOLOv8-Assisted Knee Osteoarthritis Severity Assessment from X-ray Images

Open Access
Published: 21 July, 2026
Pages: 210-223
  • Dhanya K.R. Dhanya K.R.
    Assistant Professor, Department of Computer Science with Artificial Intelligence and Data Science, KPR College of Arts, Science and Research, Coimbatore, India
    Assistant Professor, Department of Computer Science with Artificial Intelligence and Data Science, KPR College of Arts, Science and Research, Coimbatore, India
view PDF
view PDF

How to Cite

K.R., Dhanya. 2026. “YOLOv8-Assisted Knee Osteoarthritis Severity Assessment from X-Ray Images”. Journal of Artificial Intelligence and Capsule Networks 8 (3): 210-23. https://doi.org/10.36548/jaicn.2026.3.004.
ACM ACS APA ABNT Chicago Harvard IEEE MLA Turabian Vancouver AMA

Download Citation

Endnote/Zotero/Mendeley (RIS) BibTeX

Keywords

Knee Osteoarthritis
YOLOv8
Deep Learning
X-ray Image Analysis
Medical Image Classification

Abstract

Knee osteoarthritis is a progressive degenerative joint disorder characterized by cartilage deterioration, leading to pain, stiffness, reduced mobility, and functional impairment. Early detection and accurate grading of disease severity play a crucial role in effective clinical decision-making and prevention of further joint damage. Most existing deep learning approaches focus mainly on image-level classification using conventional CNN architectures and do not provide precise localization of the affected knee region, which is essential for clinical interpretation. In addition, limited research has explored unified object detection frameworks for simultaneous localization and severity grading of knee osteoarthritis. To address these limitations, this study proposes a YOLOv8-based deep learning framework for automated knee osteoarthritis detection and severity classification using radiographic X-ray images collected from a publicly available Kaggle dataset. The dataset is carefully annotated using Label Studio by marking the knee joint region and assigning five severity grades: Healthy, Doubtful, Minimal, Moderate, and Severe. The proposed model is trained using GPU acceleration to enhance computational efficiency and improve detection performance. Experimental evaluation is performed using standard metrics and the proposed system achieves a mAP@0.5 of 85.2%, demonstrating strong capability in both accurate localization and classification of knee osteoarthritis severity. It can be deployed in clinical decision support systems, radiology departments, and telemedicine platforms to enable rapid knee X-ray screening, providing radiologists, orthopaedic specialists, and physicians with an automated second opinion that reduces diagnostic time and human variability.

References

  1. Steinmetz, Jaimie D., Garland T. Culbreth, Lydia M. Haile, Quinn Rafferty, Justin Lo, Kai Glenn Fukutaki, Jessica A. Cruz et al. "Global, Regional, and National Burden of Osteoarthritis, 1990–2020 and Projections to 2050: A Systematic Analysis for the Global Burden of Disease Study 2021." The Lancet Rheumatology 2023, vol. 5, no. 9: e508-e522.
  2. Thomas, Kevin A., Łukasz Kidziński, Eni Halilaj, Scott L. Fleming, Guhan R. Venkataraman, Edwin HG Oei, Garry E. Gold, and Scott L. Delp. "Automated Classification of Radiographic Knee Osteoarthritis Severity Using Deep Neural Networks." Radiology: Artificial Intelligence 2020, vol. 2, no. 2: e190065.
  3. Jin, Ziyi, Dandan Wang, Huayong Zhang, Jun Liang, Xuebing Feng, Jinkou Zhao, and Lingyun Sun. "Incidence Trend of Five Common Musculoskeletal Disorders from 1990 to 2017 at the Global, Regional and National Level: Results from the Global Burden of Disease Study 2017." Annals of the rheumatic diseases 2020, vol. 79, no. 8: 1014-1022.
  4. Tiulpin, Aleksei, and Simo Saarakkala. "Automatic grading of Individual Knee Osteoarthritis Features in Plain Radiographs Using Deep Convolutional Neural Networks." Diagnostics 2020, vol. 10, no. 11, 932.
  5. Yunus, Usman, Javeria Amin, Muhammad Sharif, Mussarat Yasmin, Seifedine Kadry, and Sujatha Krishnamoorthy. "Recognition of knee Osteoarthritis (KOA) Using YOLOv2 and classification Based on Convolutional Neural Network." Life 2022, vol. 12, no. 8: 1126.
  6. Chen, Pingjun, Linlin Gao, Xiaoshuang Shi, Kyle Allen, and Lin Yang. "Fully Automatic Knee Osteoarthritis Severity Grading Using Deep Neural Networks with a Novel Ordinal Loss." Computerized Medical Imaging and Graphics 2019, vol. 75, 84-92.
  7. Tiulpin, Aleksei, Jérôme Thevenot, Esa Rahtu, Petri Lehenkari, and Simo Saarakkala. "Automatic Knee Osteoarthritis Diagnosis from Plain Radiographs: A Deep Learning-Based Approach." Scientific reports 2018, vol. 8, no. 1, 1727.
  8. Ren, Xiaolu, Lingxuan Hou, Shan Liu, Peng Wu, Siming Liang, Haitian Fu, Chengquan Li, Ting Li, and Yongjing Cheng. "OA-MEN: A Fusion Deep Learning Approach for Enhanced Accuracy in Knee Osteoarthritis Detection and Classification Using X-Ray Imaging." Frontiers in Bioengineering and Biotechnology 2025, vol. 12, 1437188.
  9. Pi, Sun-Woo, Byoung-Dai Lee, Mu Sook Lee, and Hae Jeong Lee. "Ensemble Deep-Learning Networks for Automated Osteoarthritis Grading in Knee X-ray Images." Scientific Reports 2023, vol. 13, no. 1, 22887.
  10. Shahid, Saman, Aamir Wali, Aatir Javaid, Shahid Zikria, Onur Osman, and Jawad Rasheed. "Potential of AI-based Diagnostic Grading System for Knee Osteoarthritis." Frontiers in Medicine 2025, vol. 12, 1707588.
  11. Cueva, Joseph Humberto, Darwin Castillo, Héctor Espinós-Morató, David Durán, Patricia Díaz, and Vasudevan Lakshminarayanan. "Detection and Classification of Knee Osteoarthritis." Diagnostics 2022, vol. 12, no. 10, 2362.
  12. Zhong, Guoyun, Peng Ding, Jianfeng He, Meifeng Liu, Haibo Sun, Fan Diao, Xueyuan Wang, Lin Ma, and Mengmeng Wang. "KOA-YOLOv8s: An Automatic Diagnostic Algorithm for X-ray Images of Knee Osteoarthritis Based on the Improved YOLOv8s." Journal of King Saud University Computer and Information Sciences 2026, vol. 38, no. 4, 140.