Classification of Knee Osteoarthritis using CNN
Creators
- 1. Information and Language Processing Research Lab, Department of Computer Science and Engineering, Kathmandu University, Dhulikhel, Kavre, Nepal
Description
Knee osteoarthritis (OA) is a joint disease which is globally common in elder people. It is typically the result of wear and tear and progressive loss of articular cartilage. It has no cure. Despite of its high prevalence, there is a lack of diagnostic tools and approaches that detects and classifies the different stages of Knee OA severalties with better precision. This paper presents the approaches to automatically quantify the severity of knee OA using X-ray images. Two of the CNN classifiers namely, VGG-15 and ResNet-32 have been used for classifying the knee OA severity into one of the 5 Kellgren-Lawrence classification grades (normal, doubtful, mild, moderate and severe). These models have been trained using loss function: ‘categorical cross entropy’ and optimizer ‘Adam’. The datasets used in this work has been collected from Bhaktapur Hospital. About 350 X-ray images were collected and manually classified into their KL grades and then they were used for testing as well as training the models. The test results shows that the accuracy of classifying knee OA severities with VGG-16 and ResNet-32 were 59% and 57% respectively. It seemed that the accuracy of VGG-16 model is better than ResNet-32 in quantifying knee OA severity.
Files
Classification of Knee Osteoarthritis 2 -Formatted Paper New.pdf
Files
(681.8 kB)
| Name | Size | Download all |
|---|---|---|
|
md5:576a0701f8c5fed17dd0fa30b81502ba
|
681.8 kB | Preview Download |
Additional details
References
- Kellgren, J. H., & Lawrence, J. (1957). Radiological assessment of osteo-arthrosis. Annals of the rheumatic diseases, 16(4), 494.
- Behiels, G., Vandermeulen, D., Maes, F., Suetens, P., & Dewaele, P. (1999, September). Active shape modelbased segmentation of digital X-ray images. In International Conference on Medical Image Computing and Computer-Assisted Intervention (pp. 128-137). Springer, Berlin, Heidelberg.
- St Clair SF, Higuera C, Krebs V, Tadross NA, Dumpe J, Barsoum (2006). Hip and Knee Arthroplasty in the Geriatric Population.Clin Geriatr Med 2006;3:515–533. [PubMed: 16860243]
- Shamir, L., Ling, S. M., Scott, W. W., Bos, A., Orlov, N., Macura, T. J., ... & Goldberg, I. G. (2008). Knee x-ray image analysis method for automated detection of osteoarthritis. IEEE Transactions on Biomedical Engineering, 56(2), 407-415..
- Bandyopadhyay, S. K. (2011). An edge detection algorithm for human knee osteoarthritis images. Journal of Global Research in Computer Science, 2(4).
- Deokar, D. D., & Patil, C. G. (2015). Effective feature extraction based automatic knee osteoarthritis detection and classification using neural network. International Journal of Engineering and Techniques, ISSN, 2395-1303.
- He, K., Zhang, X., Ren, S., & Sun, J. (2016). Deep residual learning for image recognition. In Proceedings of the IEEE conference on computer vision and pattern recognition (pp. 770-778).
- Targ, S., Almeida, D., & Lyman, K. (2016). Resnet in resnet: Generalizing residual architectures. arXiv preprint arXiv:1603.08029.
- Lazzarini, N., Runhaar, J., Bay- Jensen, A. C., Thudium, C. S., Bierma-Zeinstra, S. M. A., Henrotin, Y., & Bacardit, J. (2017). A machine learning approach for the identification of new biomarkers for knee osteoarthritis development in overweight and obese women. Osteoarthritis and cartilage, 25(12), 2014-2021.
- Antony, J., McGuinness, K., Moran, K., & O'Connor, N. E. (2017, July). Automatic detection of knee joints and quantification of knee osteoarthritis severity using convolutional neural networks. In International conference on machine learning and data mining in pattern recognition (pp. 376-390). Springer, Cham.