Knee osteoarthritis classification in X-Ray images using Ensemble Learning

• Ana Carolina Manso Silvério PUC Minas
• Alexei Manso Correa Machado PUC Minas / UFMG

DOI: https://doi.org/10.5753/sbcas.2025.6953

Abstract

Knee osteoarthritis (OA) is an inflammatory disease that can cause deformity of the articular cartilage, among other symptoms, and whose accurate diagnosis is essential to predict its progression. This work proposes an ensemble of deep neural networks such as ResNet50, Xception, Inception ResNetV2 and EfficientNet to classify OA from X-ray images, according to the levels established by the Kellgren Lawrence scale. A preliminary analysis indicates that the ensemble model surpasses results obtained by individual networks, providing a more assertive classification of the different stages of the disease.

References

Brejnebøl, M., Hansen, P., J., N., Bachmann, R., Ratjen, U., Hansen, I., Lenskjold, A., Axelsen, M., Lundemann, M., and Boesen, M. (2022). External validation of an artificial intelligence tool for radiographic knee osteoarthritis severity classification. European Journal of Radiology, 150:110249.

Chen, P., Gao, L., Shi, X., Allen, K., and L., Y. (2019). Fully automatic knee osteoarthritis severity grading using deep neural networks with a novel ordinal loss. COMPUT MED IMAG GRAP, 75:84–92.

Farajzadeh, N., Sadeghzadeh, N., and Hashemzadeh, M. (2023). Ijes-oa net: A residual neural network to classify knee osteoarthritis from radiographic images based on the edges of the intra-joint spaces. MED ENG PHYS, 113:103957.

Kishore, V., Kalpana, V., and Kumar, H. (2023). Evaluating the efficacy of deep learning models for knee osteoarthritis prediction based on kellgren-lawrence grading system. e-Prime - Advances in Electrical Engineering, Electronics and Energy, 5:100266.

Kohn, M., Sassoon, A., and Fernando, D. (2016). Classifications in brief: Kellgren-lawrence classification of osteoarthritis. CLIN ORTHOP RELAT R, 474(8):1886–1893.

Kwon, S., Han, H., Lee, M., Kim, H., Ku, Y., and Ro, D. (2020). Machine learning-based automatic classification of knee osteoarthritis severity using gait data and radiographic images. IEEE ACCESS, 8:120597–603.

Liu, B., Luo, J., and Huang, H. (2020). Toward automatic quantification of knee osteoarthritis severity using improved faster r-cnn. INT J COMPUT ASS RAD, 15:457–466.

Neslihan, B., Miika, N., and Simo, S. (2022). Machine learning based texture analysis of patella from x-rays for detecting patellofemoral osteoarthritis. INT J MED INFORM, 157:104627–104627.

Norman, B., Pedoia, V., Noworolski, A., Noworolski, A., Link, T., and Majumdar, S. (2019). Applying densely connected convolutional neural networks for staging osteoarthritis severity from plain radiographs. J DIGIT IMAGING, 32:471–477.

Sarvamangala, D. and Kulkarni, V. (2021). Grading of knee osteoarthritis using convolutional neural networks. NEURAL PROCESS LETT, 53(4):2985–3009.

Schwartz, A., Clarke, H., Spangehl, M., Bingham, J., Etzioni, D., and Neville, M. (2020). Can a convolutional neural network classify knee osteoarthritis on plain radiographs as accurately as fellowship-trained knee arthroplasty surgeons? J ARTHROPLASTY, 35:2423–2428.