Three-Dimensional Convolutional Neural Networks for Automated Knee Osteoarthritis Grading
Main Article Content
Abstract
Osteoarthritis is the most prevalent form of arthritis and frequently affects the knee joint. While deep learning methods have been widely applied to medical image analysis, three-dimensional models remain underexplored for knee osteoarthritis assessment. In this study, we employ a 3D convolutional neural network (CNN) to analyze knee magnetic resonance imaging sequences and classify disease severity according to the Kellgren–Lawrence grading system (grades 0–4). By capturing spatial correlations across image slices, the proposed approach demonstrates improved classification effectiveness compared with the 2D CNN. These findings highlight the potential of combining volumetric magnetic resonance imaging with 3D CNN to support more accurate knee osteoarthritis evaluation in clinical practice.
Keywords
3D CNN, Knee Osteoarthritis, 3D MRI.
Article Details
References
[2] M. Fransen, S. McConnell, A. R. Harmer, M. Van der Esch, M. Simic, and K. L. Bennell, “Exercise for osteoarthritis of the knee: a Cochrane systematic review,” Br. J. Sports Med., 49(24):1554-1557, 2015, DOI:10.1136/bjsports-2015-095424.
[3] D. T. Felson, A. Naimark, J. Anderson, L. Kazis, W. Castelli, and R. F. Meenan, “The prevalence of knee osteoarthritis in the elderly. The Framingham Osteoarthritis Study,” Arthritis Rheum. Off. J. Am. Coll. Rheumatol., 30(8):914-918, 1987, DOI:10.1002/art.1780300811.
[4] G. Jones, C. Ding, F. Scott, M. Glisson, and F. Cicuttini, “Early radiographic osteoarthritis is associated with substantial changes in cartilage volume and tibial bone surface area in both males and females,” Osteoarthritis Cartilage,12(2): 169-174, 2004, DOI:10.1016/j.joca.2003.08.010.
[5] D. Bhatia, T. Bejarano, and M. Novo, “Current interventions in the management of knee osteoarthritis,” J. Pharm. Bioallied Sci., 5(1):30-38, 2013, DOI:10.4103/0975-7406.106561.
[6] F. W. Roemer, C. K. Kwoh, D. Hayashi, D. T. Felson, and A. Guermazi, “The role of radiography and MRI for eligibility assessment in DMOAD trials of knee OA,” Nat. Rev. Rheumatol., 14(6):372-380, 2018, DOI:10.1038/s41584-018-0010-z.
[7] V. Juras, G. Chang, and R. R. Regatte, “Current status of functional MRI of osteoarthritis for diagnosis and prognosis,” Curr. Opin. Rheumatol., 32(1)102-109, 2020, DOI:10.1097/BOR.0000000000000674.
[8] J. Antony, K. McGuinness, N. E. O’Connor, and K. Moran, “Quantifying radiographic knee osteoarthritis severity using deep convolutional neural networks,” in 2016 23rd international conference on pattern recognition (ICPR), IEEE, 2016, pp.1195-1200.
[9] 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 International conference on machine learning and data mining in pattern recognition, Springer, 2017, pp.376-390.
[10] M. Górriz, J. Antony, K. McGuinness, X. Giró-i-Nieto, and N. E. O’Connor, “Assessing knee OA severity with CNN attention-based end-to-end architectures,” in International conference on medical imaging with deep learning, PMLR, 2019, pp.197-214.
[11] R. T. Wahyuningrum, L. Anifah, I. K. E. Purnama, and M. H. Purnomo, “A new approach to classify knee osteoarthritis severity from radiographic images based on CNN-LSTM method,” in 2019 IEEE 10th International Conference on Awareness Science and Technology (iCAST), IEEE, 2019, pp.1-6.
[12] J. H. Kellgren and J. Lawrence, “Radiological assessment of osteo-arthrosis,” Ann Rheum Dis, 16(4):494-502, 1957, https://doi.org/10.1136/ard.16.4.494.
[13] T. Wang, K. Leung, K. Cho, G. Chang, and C. M. Deniz, “Total knee replacement prediction using structural MRIs and 3D convolutional neural networks,” in International Conference on Medical Imaging with Deep Learning (MIDL) - Extended Abstract Track, 2019, pp.1-5.