Deep Learning-Based Knee Osteoarthritis Severity Classification from X-Ray Images Using Transfer Learning and Web-Based Decision Support

This paper presented a deep learning-based knee os-teoarthritis severity classification system using transfer learn-ing and web-based decision support. The system classifies knee X-ray images into five Kellgren-Lawrence grades using a MobileNetV2-based model. The training pipeline includes augmentation, class weighting, focal loss, checkpointing, and validation monitoring. The system also provides confidence-aware prediction, class probability output, and rule-based progression guidance. A Flask web application was developed to support X-ray upload, prediction result viewing, and prediction history ac-cess. Testing on 1,656 images achieved an accuracy of 55.31% and a macro F1-score of 52.12%. The model performs better for Healthy and Severe Osteoarthritis classes, while adjacent grades remain difficult. The system is suitable as an academic prototype and educational assistive tool, but further clinical validation and performance improvement are necessary before practical medical deployment.

[1]     J. H. Kellgren and J. S. Lawrence, “Radiological assessment of osteo-arthrosis,” Annals of the Rheumatic Diseases, vol. 16, no. 4, pp. 494–502, 1957. [2]     Y. LeCun, Y. Bengio, and G. Hinton, “Deep learning,” Nature, vol. 521, pp. 436–444, 2015. [3]     G. Litjens et al., “A survey on deep learning in medical image analysis,” Medical Image Analysis, vol. 42, pp. 60–88, 2017. [4]     S. J. Pan and Q. Yang, “A survey on transfer learning,” IEEE Transac-tions on Knowledge and Data Engineering, vol. 22, no. 10, pp. 1345–1359, 2010. [5]     M. Sandler, A. Howard, M. Zhu, A. Zhmoginov, and L. Chen, “Mo-bileNetV2: Inverted residuals and linear bottlenecks,” in Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, 2018, pp. 4510–4520. [6]     M. Tan and Q. V. Le, “EfficientNet: Rethinking model scaling for convolutional neural networks,” in Proceedings of the International Conference on Machine Learning, 2019, pp. 6105–6114. [7]     T. Lin, P. Goyal, R. Girshick, K. He, and P. Dolla´r, “Focal loss for dense object detection,” in Proceedings of the IEEE International Conference on Computer Vision, 2017, pp. 2980–2988. [8]     J. Antony, K. McGuinness, K. Moran, and N. E. O’Connor, “Automatic detection of knee joints and quantification of knee osteoarthritis severity using convolutional neural networks,” in Proceedings of the Interna-tional Conference on Machine Learning and Data Mining in Pattern Recognition, 2017, pp. 376–390. [9]     A. Tiulpin, J. Thevenot, E. Rahtu, P. Lehenkari, and S. Saarakkala, “Automatic knee osteoarthritis diagnosis from plain radiographs: A deep learning-based approach,” Scientific Reports, vol. 8, no. 1, pp. 1–10, 2018. [10]        P. Chen, L. Gao, X. Shi, K. Allen, and L. Yang, “Fully automatic knee osteoarthritis severity grading using deep neural networks with a novel ordinal loss,” Computerized Medical Imaging and Graphics, vol. 75, pp. 84–92, 2019.