GÖRÜNTÜ İŞLEME TABANLI ARTIK AĞLARLA DİZ OSTEOARTRİTİ TESPİTİ: SINIF TABANLI ÖZELLİK SEÇİMİ VE FÜZYONU

Hatice Akköse; Mesut Toğaçar

TR EN

GÖRÜNTÜ İŞLEME TABANLI ARTIK AĞLARLA DİZ OSTEOARTRİTİ TESPİTİ: SINIF TABANLI ÖZELLİK SEÇİMİ VE FÜZYONU

Abstract

Diz osteoartriti, eklem kıkırdak yapısının zamanla bozulması sonucunda ortaya çıkan, ağrı, hareket kısıtlılığı ve yaşam kalitesinde belirgin düşüşe yol açan kronik bir eklem hastalığıdır. Hastalığın şiddeti, klinik çalışmalarda yaygın olarak kullanılan Kellgren–Lawrence derecelendirme sistemiyle (0: normal, 1–4: artan şiddet düzeyleri) değerlendirilmektedir. Son yıllarda yapay zekâ ve derin öğrenme tabanlı yaklaşımlar, diz osteoartritinin erken ve doğru tanısında önemli başarılar göstermiştir. Bu çalışmada, diz osteoartriti şiddetinin sınıflandırılmasına yönelik ön işlem, model eğitimi, son işlem ve sınıflandırma olmak üzere dört aşamadan oluşan hibrit bir yöntem önerilmiştir. Ön işlem aşamasında orijinal X-ışını görüntüleri LBP ve Grad-CAM teknikleriyle işlenmiş ve üç ayrı veri kümesi elde edilmiştir. Model eğitimi aşamasında bu veri kümeleri ResNet-18 modeliyle eğitilerek her bir veri kümesi için sınıf tabanlı özellik setleri elde edilmiştir. Son işlem aşamasında elde edilen bu özellik setleri özellik füzyonu yöntemiyle birleştirilmiş ve performansları SVM yöntemiyle analiz edilmiştir. En iyi başarı gösteren özellik seti için mRMR ve Relief özellik seçim algoritmaları ayrı ayrı uygulanmış, her birinden seçilen 10 özelliğin içerisinden tekrar eden özellikler çıkarılarak 13 özellikli nihai özellik seti elde edilmiştir. Sınıflandırma aşamasında nihai özellik seti, SVM algoritmasıyla sınıflandırılmış ve %99,8 doğruluk oranına ulaşılmıştır.

Keywords

KNEE OSTEOARTHRITIS DETECTION WITH IMAGE PROCESSING-BASED RESIDUAL NETWORKS: CLASS-BASED FEATURE SELECTION AND FUSION

Abstract

Knee osteoarthritis is a chronic joint disease that results from the gradual deterioration of joint cartilage structure over time, leading to pain, limited mobility, and a significant decline in quality of life. The severity of the disease is assessed using the Kellgren–Lawrence grading system (0: normal, 1–4: increasing severity levels), which is widely used in clinical studies. In recent years, artificial intelligence and deep learning-based approaches have shown significant success in the early and accurate diagnosis of knee osteoarthritis. This study proposes a hybrid method consisting of four stages: preprocessing, model training, postprocessing, and classification for the classification of knee osteoarthritis severity. In the preprocessing stage, the original X-ray images were processed using LBP and Grad-CAM techniques, yielding three separate datasets. In the model training stage, these datasets were trained using the ResNet-18 model, yielding class-based feature sets for each dataset. In the postprocessing stage, these feature sets were combined using a feature fusion method, and their performance was analyzed using the SVM method. For the feature set showing the best performance, the mRMR and Relief feature selection algorithms were applied separately, and redundant features were removed from the 10 features selected from each, resulting in a final feature set of 13 features. In the classification phase, the final feature set was classified using the SVM algorithm, achieving an accuracy rate of 99.8%.

Keywords

References

Ahmed, S. M., & Mstafa, R. J. (2022). Identifying Severity Grading of Knee Osteoarthritis from X-ray Images Using an Efficient Mixture of Deep Learning and Machine Learning Models. Diagnostics, 12(12). https://doi.org/10.3390/diagnostics12122939
Ali, H. M. (2025). Osteoarthritis Classification Algorithm Using CNN and Image Edge Detections. Journal of Engineering Sciences, 53(1), 102-117. https://doi.org/10.21608/jesaun.2024.266440.1306
Alruwaili, M., Ali, A., Almutairi, M., Alsahyan, A., & Mohamed, M. (2025). LSTM and ResNet18 for optimized ambulance routing and traffic signal control in emergency situations. Scientific reports, 15(1), 6011. https://doi.org/10.1038/s41598-025-89651-4
Armaghani, D. J., Asteris, P. G., Askarian, B., Hasanipanah, M., Tarinejad, R., & Huynh, V. Van. (2020). Examining hybrid and single SVM models with different kernels to predict rock brittleness. Sustainability (Switzerland), 12(6), 1-17. https://doi.org/10.3390/su12062229
Chen, P. (2018). Knee Osteoarthritis Severity Grading Dataset. 1. https://doi.org/10.17632/56RMX5BJCR.1
Chen, P., Gao, L., Shi, X., Allen, K., & Yang, L. (2019). Fully automatic knee osteoarthritis severity grading using deep neural networks with a novel ordinal loss. Computerized Medical Imaging and Graphics, 75, 84-92. https://doi.org/10.1016/j.compmedimag.2019.06.002
Culvenor, A. G., Engen, C. N., Øiestad, B. E., Engebretsen, L., & Risberg, M. A. (2015). Defining the presence of radiographic knee osteoarthritis: a comparison between the Kellgren and Lawrence system and OARSI atlas criteria. Knee Surgery, Sports Traumatology, Arthroscopy, 23(12), 3532-3539. https://doi.org/10.1007/S00167-014-3205-0
Singh, D. DLOA (Part-18)-ResNet CNN and Implementation. (2025). https://medium.com/@singhdewansh99/dlao-part-18-resnet-cnn-and-implementation-e3ce3d719807 Accessed 01.06.25.

Erdoğan, F., Tongur, V., & Uzbaş, B. (2024). VERİ ÖN İŞLEME TEKNİKLERİNİN SAĞLIK VERİLERİNİN SINIFLANDIRMA BAŞARISINA ETKİSİNİN İNCELENMESİ. Adıyaman Üniversitesi Mühendislik Bilimleri Dergisi, 11(24), 475-488. https://doi.org/10.54365/adyumbd.1466631
He, K., Zhang, X., Ren, S., & Sun, J. (2015). Deep Residual Learning for Image Recognition. https://doi.org/10.48550/arXiv.1512.03385
Jahan, M., Hasan, M. Z., Jahan Samia, I., Fatema, K., Rony, M. A. H., Shamsul Arefin, M., & Moustafa, A. (2024). KOA-CCTNet: An Enhanced Knee Osteoarthritis Grade Assessment Framework Using Modified Compact Convolutional Transformer Model. IEEE Access, 12, 107719-107741. https://doi.org/10.1109/ACCESS.2024.3435572
Keerthana, D., Venugopal, V., Nath, M. K., & Mishra, M. (2023). Hybrid convolutional neural networks with SVM classifier for classification of skin cancer. Biomedical Engineering Advances, 5, 100069. https://doi.org/10.1016/j.bea.2022.100069
Lee, L. S., Chan, P. K., Fung, W. C., Chan, V. W. K., Yan, C. H., & Chiu, K. Y. (2021). Imaging of knee osteoarthritis: A review of current evidence and clinical guidelines. Musculoskeletal Care, 19(3), 363-374. https://doi.org/10.1002/msc.1536
Li, Z., Deng, S., Hong, Y., Wei, Z., & Cai, L. (2024). A novel hybrid CNN–SVM method for lithology identification in shale reservoirs based on logging measurements. Journal of Applied Geophysics, 223. https://doi.org/10.1016/j.jappgeo.2024.105346
Liu, K.-Y., Lu, N.-H., Huang, Y.-H., Matsushima, A., Kimura, K., Okamoto, T., & Chen, T.-B. (2025). Majority Voting Ensemble of Deep CNNs for Robust MRI-Based Brain Tumor Classification. Diagnostics, 15(14), 1782. https://doi.org/10.3390/diagnostics15141782
Mahfus, M., Tosun, M., Göker, H., & Kelimeler Öz, A. (2025). TRANSFER LEARNING-BASED CLASSIFICATION OF KNEE OSTEOARTHRITIS SEVERITY FROM X-RAY IMAGES DIZ OSTEOARTRITI ŞIDDETININ X-RAY GÖRÜNTÜLERINDEN TRANSFER ÖĞRENME TABANLI SINIFLANDIRILMASI. Mühendislik Bilimleri ve Tasarım Dergisi, 13(1), 325-339. https://doi.org/10.21923/jesd.1608509
Mohammed, A. S., Hasanaath, A. A., Latif, G., & Bashar, A. (2023). Knee Osteoarthritis Detection and Severity Classification Using Residual Neural Networks on Preprocessed X-ray Images. Diagnostics, 13(8), 1380. https://doi.org/10.3390/DIAGNOSTICS13081380
Ou, X., Yan, P., Zhang, Y., Tu, B., Zhang, G., Wu, J., & Li, W. (2019). Moving Object Detection Method via ResNet-18 with Encoder-Decoder Structure in Complex Scenes. IEEE Access, 7, 108152-108160. https://doi.org/10.1109/ACCESS.2019.2931922
Pi, S. W., Lee, B. D., Lee, M. S., & Lee, H. J. (2023). Ensemble deep-learning networks for automated osteoarthritis grading in knee X-ray images. Scientific Reports, 13(1). https://doi.org/10.1038/s41598-023-50210-4
Ren, J. Le, Yang, J., & Hu, W. (2025). The global burden of osteoarthritis knee: a secondary data analysis of a population-based study. Clinical Rheumatology, 44(4), 1769-1810. https://doi.org/10.1007/S10067-025-07347-6/TABLES/6
Rizki, I., Subekti, S., Indriyanto, S., Rizal, A., Triwiyanto, T., & Ziani, S. (2024). Epileptic Seizure Detection using DWT based on MRMR Feature Selection Method. Proceedings - IEIT 2024 - 2024 International Conference on Electrical and Information Technology, 84-89. https://doi.org/10.1109/IEIT64341.2024.10763220
Rohani, M. M. (t.y.). Automated Knee Osteoarthritis Severity Grading in X-ray Images Using Transfer Learning. Içinde International Journal of Applied Data Science in Engineering and Health (C. 1). https://ijadseh.com/index.php/ijadseh/article/view/33
Sedaghatjoo, Z., Hosseinzadeh, H., & Bigham, B. S. (2024). Local Binary Pattern(LBP) Optimization for Feature Extraction. https://doi.org/10.48550/arXiv.2407.18665
Selvaraju, R. R., Das, A., Vedantam, R., Cogswell, M., Parikh, D., & Batra, D. (2016). Grad-CAM: Why did you say that? https://doi.org/10.48550/arXiv.1611.07450
Sharma, G., Anand, V., & Kumar, V. (2023). Classification of Osteo-Arthritis with the Help of Deep Learning and Transfer Learning. Proceedings of the 5th International Conference on Inventive Research in Computing Applications, ICIRCA 2023, 446-452. Institute of Electrical and Electronics Engineers Inc. https://doi.org/10.1109/ICIRCA57980.2023.10220816
Sohail, M., Azad, M. M., & Kim, H. S. (2025). Knee osteoarthritis severity detection using deep inception transfer learning. Computers in Biology and Medicine, 186, 109641. https://doi.org/10.1016/J.COMPBIOMED.2024.109641
Thomas, K. A., Kidziński, Ł., Halilaj, E., Fleming, S. L., Venkataraman, G. R., Oei, E. H. G., … Delp, S. L. (2020). Automated classification of radiographic knee osteoarthritis severity using deep neural networks. Radiology: Artificial Intelligence, 2(2). https://doi.org/10.1148/RYAI.2020190065
Touahema, S., Zaimi, I., Zrira, N., Ngote, M. N., Doulhousne, H., & Aouial, M. (2024). MedKnee: A New Deep Learning-Based Software for Automated Prediction of Radiographic Knee Osteoarthritis. Diagnostics, 14(10). https://doi.org/10.3390/diagnostics14100993
Vad, V. B., Adin, D. R., & Solomon, J. (2023). Knee Osteoarthritis. Critical Reviews in Physical and Rehabilitation Medicine, 16(3), 211-231. https://doi.org/10.1615/CritRevPhysRehabilMed.v16.i3.30
Ye, Z., Peng, J., Zhang, X., & Song, L. (2024). Identification of OSAHS patients based on ReliefF-mRMR feature selection. Physical and Engineering Sciences in Medicine, 47(1), 99-108. https://doi.org/10.1007/s13246-023-01345-1
Zehra Solak, F. (2024). Karaelmas Science and Engineering Journal Classification of Knee Osteoarthritis Severity by Transfer Learning from X-Ray Images X-Ray Görüntülerinden Transfer Öğrenme ile Diz Osteoartriti Şiddetinin Sınıflandırılması. Karaelmas Fen ve Müh. Derg, 14(2), 119-133. https://doi.org/10.7212/karaelmasfen.1480055

Details

Primary Language

Turkish

Subjects

Image Processing, Deep Learning

Journal Section

Research Article

Authors

Hatice Akköse ^*
0009-0004-2041-5569
Türkiye

Mesut Toğaçar
0000-0002-8264-3899
Türkiye

Publication Date

June 3, 2026

Submission Date

October 15, 2025

Acceptance Date

March 2, 2026

Published in Issue

Year 2026 Volume: 29 Number: 2

IZ

https://izlik.org/JA77JU44YB

Cite

RIS / Bibtex

APA

Akköse, H., & Toğaçar, M. (2026). GÖRÜNTÜ İŞLEME TABANLI ARTIK AĞLARLA DİZ OSTEOARTRİTİ TESPİTİ: SINIF TABANLI ÖZELLİK SEÇİMİ VE FÜZYONU. Kahramanmaraş Sütçü İmam Üniversitesi Mühendislik Bilimleri Dergisi, 29(2), 592-607. https://izlik.org/JA77JU44YB

AMA

1.Akköse H, Toğaçar M. GÖRÜNTÜ İŞLEME TABANLI ARTIK AĞLARLA DİZ OSTEOARTRİTİ TESPİTİ: SINIF TABANLI ÖZELLİK SEÇİMİ VE FÜZYONU. KSU J. Eng. Sci. 2026;29(2):592-607. https://izlik.org/JA77JU44YB

Chicago

Akköse, Hatice, and Mesut Toğaçar. 2026. “GÖRÜNTÜ İŞLEME TABANLI ARTIK AĞLARLA DİZ OSTEOARTRİTİ TESPİTİ: SINIF TABANLI ÖZELLİK SEÇİMİ VE FÜZYONU”. Kahramanmaraş Sütçü İmam Üniversitesi Mühendislik Bilimleri Dergisi 29 (2): 592-607. https://izlik.org/JA77JU44YB.

EndNote

Akköse H, Toğaçar M (June 1, 2026) GÖRÜNTÜ İŞLEME TABANLI ARTIK AĞLARLA DİZ OSTEOARTRİTİ TESPİTİ: SINIF TABANLI ÖZELLİK SEÇİMİ VE FÜZYONU. Kahramanmaraş Sütçü İmam Üniversitesi Mühendislik Bilimleri Dergisi 29 2 592–607.

IEEE

[1]H. Akköse and M. Toğaçar, “GÖRÜNTÜ İŞLEME TABANLI ARTIK AĞLARLA DİZ OSTEOARTRİTİ TESPİTİ: SINIF TABANLI ÖZELLİK SEÇİMİ VE FÜZYONU”, KSU J. Eng. Sci., vol. 29, no. 2, pp. 592–607, June 2026, [Online]. Available: https://izlik.org/JA77JU44YB

ISNAD

Akköse, Hatice - Toğaçar, Mesut. “GÖRÜNTÜ İŞLEME TABANLI ARTIK AĞLARLA DİZ OSTEOARTRİTİ TESPİTİ: SINIF TABANLI ÖZELLİK SEÇİMİ VE FÜZYONU”. Kahramanmaraş Sütçü İmam Üniversitesi Mühendislik Bilimleri Dergisi 29/2 (June 1, 2026): 592-607. https://izlik.org/JA77JU44YB.

JAMA

1.Akköse H, Toğaçar M. GÖRÜNTÜ İŞLEME TABANLI ARTIK AĞLARLA DİZ OSTEOARTRİTİ TESPİTİ: SINIF TABANLI ÖZELLİK SEÇİMİ VE FÜZYONU. KSU J. Eng. Sci. 2026;29:592–607.

MLA

Akköse, Hatice, and Mesut Toğaçar. “GÖRÜNTÜ İŞLEME TABANLI ARTIK AĞLARLA DİZ OSTEOARTRİTİ TESPİTİ: SINIF TABANLI ÖZELLİK SEÇİMİ VE FÜZYONU”. Kahramanmaraş Sütçü İmam Üniversitesi Mühendislik Bilimleri Dergisi, vol. 29, no. 2, June 2026, pp. 592-07, https://izlik.org/JA77JU44YB.

Vancouver

1.Hatice Akköse, Mesut Toğaçar. GÖRÜNTÜ İŞLEME TABANLI ARTIK AĞLARLA DİZ OSTEOARTRİTİ TESPİTİ: SINIF TABANLI ÖZELLİK SEÇİMİ VE FÜZYONU. KSU J. Eng. Sci. [Internet]. 2026 Jun. 1;29(2):592-607. Available from: https://izlik.org/JA77JU44YB