Classification of Knee Osteoarthritis Severity by Transfer Learning from X-Ray Images

Yıl 2024, Cilt: 14 Sayı: 2, 119 - 133, 23.07.2024

Fatma Zehra Solak

Öz

Knee Osteoarthritis (KOA) is the most common type of arthritis and its severity is assessed with the Kellgren-Lawrence (KL) grading system based on evidence from both knee bones. Recent advancements point to an era where computer-assisted methods enhance KOA diagnostic efficiency. This study implemented binary and multiple classification processes based on X-ray images and deep learning algorithms for computer-aided KOA severity diagnosis. Pre-processing involved extracting the region of interest and contrast enhancement with CLAHE on the X-ray images from the included dataset. Using this dataset, 2, 3, 4, and 5 class classification processes were conducted with ResNet-50, Xception, VGG16, EfficientNetb0, and DenseNet201 transfer learning models. Each model was assessed with “rmsprop,” “sgdm,” and “adam” optimization algorithms. Study findings reveal that, the DenseNet201-rmsprop model achieved 87.7% accuracy, 87.2% F1-Score, and a 0.75 Cohen’s kappa value for 2-class classification. For 3-class classification, it achieved 85.6% accuracy, 82.4% F1-Score, and a 0.71 Cohen’s kappa value. For 4-class classification, the DenseNet201-rmsprop model provided 81.5% accuracy, 77.1% F1-Score, and a Cohen’s kappa value of 0.67. In the 5-class classification, the highest success was with the Xception-rmsprop model, with 67.8% accuracy, 68.8% F1-Score, and a 0.55 Cohen’s kappa value. The evaluation with varying class numbers and different transfer learning models highlights the proposed approach’s effectiveness. Results of the study underscore the study’s uniqueness and success in demonstrating how varying the number of classes, employing different transfer learning models and optimizers can provide clearer insights into KOA severity evaluation.

Anahtar Kelimeler

Clahe, multiple classification, osteoarthritis, transfer learning, x-ray

X-Ray Görüntülerinden Transfer Öğrenme ile Diz Osteoartriti Şiddetinin Sınıflandırılması

Öz

Knee Osteoartrit (KOA), her iki diz kemiğinden elde edilen kanıtlara dayanarak Kellgren-Lawrence (KL) derecelendirme sistemi ile değerlendirilen en yaygın artrit türüdür. Son gelişmeler, KOA tanı verimliliğini artırmak için bilgisayar destekli yöntemlerin kullanıldığı bir döneme işaret etmektedir. Bu çalışma, X-ışını görüntüleri ve derin öğrenme algoritmaları temelinde ikili ve çoklu sınıflandırma süreçleri uygulayarak KOA şiddeti tanısında bilgisayar destekli yöntemler geliştirmiştir. Önişleme işlemi, dahil edilen veri setindeki X-ışını görüntülerinden ilgi alanının çıkarılması ve kontrastın CLAHE ile arttırılmasını içermiştir. Bu veri seti kullanılarak, ResNet-50, Xception, VGG16, EfficientNetb0 ve DenseNet201 transfer öğrenme modelleri ile 2, 3, 4 ve 5 sınıf sınıflandırma süreçleri gerçekleştirilmiştir. Her model, “rmsprop,” “sgdm,” ve “adam” optimizasyon algoritmaları ile değerlendirilmiştir. Çalışmanın bulguları, DenseNet201-rmsprop modelinin 2-sınıf sınıflandırma için %87.7 doğruluk, %87.2 F1-Skoru ve 0 .75 Cohen’s kappa değeri elde ettiğini ortaya koymaktadır. 3-sınıf sınıflandırma için %85.6 doğruluk, %82.4 F1-Skoru ve 0.71 Cohen’s kappa değeri elde edilmiştir. 4-sınıf sınıflandırmada, DenseNet201-rmsprop modeli %81.5 doğruluk, %77.1 F1-Skoru ve 0.67 Cohen’s kappa değeri sağlamıştır. 5-sınıf sınıflandırmada, en yüksek başarı, %67.8 doğruluk, %68.8 F1-Skoru ve 0.55 Cohen’s kappa değeri ile Xception-rmsprop modeli ile elde edilmiştir. Farklı sınıf sayıları ve farklı aktarım öğrenme modelleri ile yapılan değerlendirme, önerilen yaklaşımın etkinliğini vurgulamaktadır. Çalışmanın sonuçları, sınıf sayısının değiştirilmesinin, farklı transfer öğrenme modellerinin ve optimize edicilerin kullanılmasının KOA şiddeti değerlendirmesinde nasıl daha net bilgiler sağlayabileceğini gösterme konusunda çalışmanın benzersizliğini ve başarısını vurgulamaktadır.

Anahtar Kelimeler

Clahe, çoklu sınıflandırma, osteoartrit, transfer öğrenme, x-ray

Kaynakça

• Ahmed, R., Imran, AS. 2024. Knee osteoarthritis analysis using deep learning and XAI on X-rays. IEEE Access, 12: 68870-68879. Doi: 10.1109/ACCESS.2024.3400987
• Alshamrani, HA., Rashid, M., Alshamrani SS., Alshehri AH. 2023. Osteo-net: An automated system for predicting knee osteoarthritis from x-ray images using transfer-learning-based neural networks approach. Healthcare, 11(9): 1206. Doi: 10.3390/healthcare11091206
• Anderson, AS., Loeser, RF. 2010. Why is osteoarthritis an age-related disease?. Best Pract Res Clin Rheumatol., 24(1): 15-26. Doi: 10.1016/j.berh.2009.08.006
• Brahim, A., Jennane, R., Riad, R., Janvier, T., Khedher, L., Toumi H., Lespessailles, E. 2019. A decision support tool for early detection of knee OsteoArthritis using X-ray imaging and machine learning: Data from the OsteoArthritis Initiative. Computerized Medical Imaging and Graphics, 73: 11-18. Doi: 10.1016/j.compmedimag.2019.01.007
• Chen, P. 2018. Knee osteoarthritis severity grading dataset. Mendeley Data, 1. Doi: 10.17632/56rmx5bjcr.1
• Chollet, F. 2017. Xception: Deep learning with depthwise separable convolutions. 2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR), p. 1800-1807, USA
• Deshpande, BR., Katz, JN., Solomon DH., Yelin, EH., Hunter, DJ., Messier, SP., Suter, LG., Losina, E. 2016. Number of persons with symptomatic knee osteoarthritis in the US: impact of race and ethnicity, age, sex, and obesity. Arthritis care & research, 68(12): 1743-1750. Doi: 10.1002/acr.22897
• Guan, B., Liu, F., Mizaian, AH., Demehri, S., Samsonov, A., Guermazi, A., Kijowski, R. 2022. Deep learning approach to predict pain progression in knee osteoarthritis. Skeletal Radiol., 51(2):363-373. Doi: 10.1007/s00256-021-03773-0
• He, K., Zhang, X., Ren, S., Sun, J. 2016. Deep residual learning for image recognition. Proceedings of the IEEE conference on computer vision and pattern recognition, p. 770-778. Doi: 10.48550/arXiv.1512.03385
• Huang, G., Liu, Z., Van Der Maaten, L., Weinberger, KQ. 2017. Densely connected convolutional networks. Proceedings of the IEEE conference on computer vision and pattern recognition. p. 4700-4708. Doi: 10.48550/arXiv.1608.06993
• Jakaite, L., Schetinin, V., Hladůvka, J., Minaev, S., Ambia, A., Krzanowski, W. 2021. Deep learning for early detection of pathological changes in x-ray bone microstructures: case of osteoarthritis. Sci. Rep., 11(1): 2294. Doi: 10.1038/s41598-021-81786-4
• Kellgren, J., Bier, F. 1956. Radiological signs of rheumatoid arthritis: a study of observer differences in the reading of hand films. Ann. Rheum. Dis., 15(1): 55–60. Doi: 10.1136/ard.15.1.55.
• Kim, DH., Kim, SC., Yoon JS., Lee YS. 2020. Are there harmful effects of preoperative mild lateral or patellofemoral degeneration on the outcomes of open wedge high tibial osteotomy for medial compartmental osteoarthritis? Orthop. J. Sports. Med., 8(6). Doi: 10.1177/2325967120927481
• kneeosteoarthritis (2018). Knee Osteoarthritis Dataset with severity grading, https://huggingface.co/datasets/SilpaCS/kneeosteoarthritis/blob/main/data.zip, Access Data: 02.02.2024
• Kohn, MD., Sassoon, AA., Fernando, ND. 2016. Classifications in brief: Kellgren-Lawrence classification of osteoarthritis. Clin. Orthop. Relat. Res., 474(8): 1886–1893. Doi: 10.1007/s11999-016-4732-4
• Kondal, S., Kulkarni, V., Gaikwad, A., Kharat, A., Pant, A. 2022. Automatic grading of knee osteoarthritis on the Kellgren-Lawrence scale from radiographs using convolutional neural networks. Advances in Deep Learning, Artificial Intelligence and Robotics: Proceedings of the 2nd International Conference on Deep Learning, Artificial Intelligence and Robotics, (ICDLAIR), p. 163-173, Springer.
• Li, W., Yu, S., Yang, R., Tian, Y., Zhu, T., Liu, H., Jiao, D., Zhang, F., Liu, X., Tao, L. 2023. Machine learning model of resnet50-ensemble voting for malignant–benign small pulmonary nodule classification on computed tomography images. Cancers, 15(22): 5417. Doi: 10.3390/cancers15225417
• Martel-Pelletier, J. (1999). Pathophysiology of osteoarthritis. Osteoarthritis and Cartilage., 7(4): 371-373. Doi: 10.1053/joca.1998.0214
• Mou, SF., Razzak, SA. 2023. Brain disease classification from MRI scans using EfficientNetB0 feature extraction. 2023 International Conference on Information and Communication Technology for Sustainable Development (ICICT4SD), IEEE. p. 336-340.
• Nasser, Y., Jennane, R., Chetouani, A., Lespessailles, E., El Hassouni, M. 2020. Discriminative Regularized Auto-Encoder for early detection of knee osteoarthritis: data from the osteoarthritis initiative. IEEE transactions on medical imaging, 39(9): 2976-2984. Doi: 10.1109/TMI.2020.2985861
• Pi, SW., Lee, BD., Lee, MS., Lee, HJ. 2023. Ensemble deep-learning networks for automated osteoarthritis grading in knee X-ray images. Scientific Reports, 13(1): 22887. Doi: 10.1038/s41598-023-50210-4
• Raza, A., Phan, TL., Li, HC., Hieu, NV., Nghia, TT., Ching, CTS. 2024. A Comparative study of machine learning classifiers for enhancing knee Osteoarthritis Diagnosis. Information, 15(4): 183. Doi: 10.3390/info15040183
• Rehman, SU., Gruhn, V. 2024. A Sequential VGG16+ CNN based Automated Approach with adaptive input for efficient detection of knee Osteoarthritis stages. IEEE Access,12: 62407 - 62415. Doi: 10.1109/ACCESS.2024.3395062
• Saini, D., Khosla, A., Chand, T., Chouhan, DK., Prakash, M. 2023. Automated knee osteoarthritis severity classification using three‐stage preprocessing method and VGG16 architecture. International Journal of Imaging Systems and Technology, 33(3): 1028-1047. Doi: 10.1002/ima.22845
• Shaheed, K., Mao, A., Qureshi, I., Kumar, M., Hussain, S., Ullah, I., Zhang, X. 2022. DS-CNN: A pre-trained Xception model based on depth-wise separable convolutional neural network for finger vein recognition. Expert Systems with Applications, 191: 116288. Doi: 10.1016/j.eswa.2021.116288
• Sharma, AK., Nandal, A., Dhaka, A., Koundal, D., Bogatinoska, DC., Alyami, H. 2022. Enhanced watershed segmentation algorithm-based modified ResNet50 model for brain tumor detection. BioMed Research International, 2022.Doi: 10.1155/2022/7348344
• Sharma, G., Anand, V., Kumar, V. 2023. Classification of Osteo-Arthritis with the help of deep learning and transfer learning. 2023 5th International Conference on Inventive Research in Computing Applications (ICIRCA), India. Doi: 10.1109/ICIRCA57980.2023.10220816
• Shivadekar, S., Kataria, B., Hundekari, S., Wanjale, K., Balpande VP., Suryawanshi, R. 2023. Deep learning based image classification of lungs radiography for detecting COVID-19 using a Deep CNN and ResNet 50. International Journal of Intelligent Systems and Applications in Engineering, 11(1s): 241-250.
• Simonyan, K., Zisserman A. 2014. Very deep convolutional networks for large-scale image recognition. arXiv preprint arXiv:1409.1556. Doi: 10.48550/arXiv.1409.1556
• Szegedy, C., Vanhoucke, V., Ioffe, S., Shlens, J., Wojna, Z. 2016. Rethinking the inception architecture for computer vision. Proceedings of the IEEE conference on computer vision and pattern recognition, p. 2818-2826. Doi:10.1109/CVPR.2016.308
• Tan, M., Le, Q. 2019. Efficientnet: Rethinking model scaling for convolutional neural networks. International conference on machine learning, p. 6105-6114, California.
• Tiulpin, A., Thevenot, J., Rahtu, E., Lehenkari, P., Saarakkala, S. 2018. Automatic knee osteoarthritis diagnosis from plain radiographs: a deep learning-based approach. Sci. Rep., 8(1): 1727. Doi: 10.1038/s41598-018-20132-7
• Turkoglu, M. 2021. COVID-19 detection system using chest CT images and multiple kernels-extreme learning machine based on deep neural network. Irbm, 42(4): 207-214. Doi: 10.1016/j.irbm.2021.01.004
• Vina, ER., Kwoh, CK. 2018. Epidemiology of osteoarthritis: literature update. Current opinion in rheumatology, 30(2):160-167. Doi: 10.1097/BOR.0000000000000479
• Wang, Y., Wang, X., Gao, T., Du, L., Liu, W. 2021. An automatic knee osteoarthritis diagnosis method based on deep learning: data from the osteoarthritis initiative. J. Healthc. Eng., 2021: 1-10. Doi: 10.1155/2021/5586529
• Wang, Y., Li, S., Zhao, B., Zhang, J., Yang, Y., Li, B. 2022. A ResNet‐based approach for accurate radiographic diagnosis of knee osteoarthritis. CAAI Transactions on Intelligence Technology, 7(3): 512-521. Doi: 10.1049/cit2.12079
• Wenham, C., Grainger, A., Conaghan, P. 2014. The role of imaging modalities in the diagnosis, differential diagnosis and clinical assessment of peripheral joint osteoarthritis. Osteoarthritis Cartilage., 22(10): 1692-1702. Doi: 10.1016/j.joca.2014.06.005
• Yang, L., Xu, S., Yu, X., Long, H., Zhang, H., Zhu, Y. 2023. A new model based on improved VGG16 for corn weed identification. Front. Plant Sci., 14. Doi: 10.3389/fpls.2023.1205151
• Yong, CW., Teo, K., Murphy, BP., Hum, YC., Tee, YK., Xia, K., Lai, KW. 2022. Knee osteoarthritis severity classification with ordinal regression module. Multimedia Tools and Applications, 81 (2): 1-13. Doi: 10.1007/s11042-021-10557-0