Keywords
texture analysis
synovial cavity
cubic SVM
kNN
machine learning
GLCM
Abstract
The article is devoted to the development of a method for automatic identification of osteoarthritis of the shoulder joint using texture analysis of X‑ray images. Osteoarthritis is a chronic degenerative disease that has a significant impact on quality of life, especially when the shoulder joint is affected. Traditional diagnostic methods often detect changes only at late stages of the disease, so early diagnosis is important. In this paper, we proposed the use of machine learning techniques such as support vector method (SVM) and k‑nearest neighbours (kNN) method to analyse radiographs. The study included radiographs of the shoulder joint of 31 patients. Image preprocessing included contrast enhancement using the CLAHE algorithm and the use of grey level matching matrix (GLCM) to extract texture features. The results showed that the kNN method showed 100 % accuracy in classifying normal images, while SVM showed 79 % accuracy. For abnormal images, both methods showed 100 % performance. The high performance of this algorithm compared to traditional methods was highlighted and the importance of further research to improve the algorithm and expand the sample was emphasised. The proposed approach allows detailed analysis of medical images, revealing subtle changes in tissues, which is especially important for early diagnosis and assessment of the extent of lesions. We believe that further research in this area is needed to improve the diagnosis and treatment of osteoarthritis of the shoulder joint.
Acknowledgments
The authors are grateful to Daria A. Akulova, Cand. Sci. (Med.), Director of the Ekaterinburg Medical Centre, for the opportunity to use the archive of X‑ray images for the study.
For citation
Zhiljakov AA, Volokitina EA. Application of radiograph texture analysis to identify signs of osteoarthritis of the shoulder joint. USMU Medical Bulletin. 2024;(2):40–52. (In Russ.). EDN: https://elibrary.ru/YVOSFP.
References
Role of Global Femoral Cartilage in Assessing Severity of Primary Knee Osteoarthritis / N. F. W. Khalil, S. El-sherif, M. M. A. El Hamid [et al.] // Egyptian Rheumatology and Rehabilitation. 2022. Vol. 49, Iss. 1, Art. No. 16. DOI: https://doi.org/10.1186/s43166-022-00115-3.
Imaging of the Acromioclavicular Joint: Anatomy, Function, Pathologic Features, and Treatment / D. V. Flores, P. K. Goes, C. M. Gómez [et al.] // RadioGraphics. 2020. Vol. 40, No. 5. P. 1355–1382. DOI: https://doi.org/10.1148/rg.2020200039.
Выбор метода использования регионарной анестезии плечевого сплетения межлестничным доступом при артроскопических операциях на плечевом суставе / С. В. Крылов, И. Н. Пасечник, М. В. Капырина, К. Ю. Уколов // Уральский медицинский журнал. 2019. № 7. С. 138–144. EDN: https://elibrary.ru/vrukpu.
Diagnostics of Articular Cartilage Damage Based on Generated Acoustic Signals Using ANN — Part II: Patellofemoral Joint / R. Karpiński, P. Krakowski, J. Jonak [et al.] // Sensors. 2022. Vol. 22, Iss. 10, Art. No. 3765. DOI: https://doi.org/10.3390/s22103765.
Diagnostics of Articular Cartilage Damage Based on Generated Acoustic Signals Using ANN — Part I: Femoral‐Tibial Joint / R. Karpiński, P. Krakowski, J. Jonak [et al.] // Sensors. 2022. Vol. 22, Iss. 6, Art. No. 2176. DOI: https://doi.org/10.3390/s22062176.
Крылов В. А. Хирургия плечевого сустава в условиях травматологического отделения многопрофильной ЦГБ малого города // Уральский медицинский журнал. 2022. Т. 21, № 2. С. 97–100. DOI: https://doi.org/10.52420/2071-5943-2022-21-2-97-100.
Association Between Pain and Radiological Changes in Patients with Knee Osteoarthritis / A. Munawar, M. M. A. Durrani, M. W. Akhtar [et al.] // Journal of Health and Rehabilitation Research. 2024. Vol. 4, Iss. 1. P. 608–612. DOI: https://doi.org/10.61919/jhrr.v4i1.406.
AB1174 Knee Symptoms are More Strongly Associated with Quadriceps Muscle Strength than Grip Strength or Muscle Mass: The Road Study / S. Muraki, T. Akune, H. Oka [et al.] // Annals of the Rheumatic Diseases. 2015. Vol. 74, Suppl. 2. P. 1295–1296. DOI: https://doi.org/10.1136/annrheumdis‑2015‑eular.2961.
LIFEx: A Freeware for Radiomic Feature Calculation in Multimodality Imaging to Accelerate Advances in the Characterization of Tumor Heterogeneity / C. Nioche, F. Orlhac, S. Boughdad [et al.] // Cancer Research. 2018. Vol. 78, Iss. 16. P. 4786–4789. DOI: https://doi.org/10.1158/0008-5472.CAN‑18-0125.
Andrade D., Gifford H., Das M. Multi-modality GLCM Image Texture Feature for Segmentation and Tissue Classification // Medical Imaging 2023 : Physics of Medical Imaging : Proceedings / Ed. by R. Fahrig, J. M. Sabol, L. Yu. SPIE, 2023. P. 124634P. Vol. 12463. DOI: https://doi.org/10.1117/12.2659078.
Husain A., Vishvakarma V. P. RES-KELM Fusion Model Based on Non-iterative Deterministic Learning Classifier for Classification of Covid19 Chest X‑Ray Images // Journal of Intelligent Systems. Vol. 32, Iss. 1, Art. No. 20220235. DOI: https://doi.org/10.1515/jisys‑2022-0235.
Лечебная физкультура при повреждении ротаторной манжеты плеча / Д. Н. Бобунов, В. Д. Михайлов, М. С. Овсянников [и др.] // Уральский медицинский журнал. 2019. № 14. С. 92–99. EDN: https://elibrary.ru/iaegcp.
Kaur P. P., Singh S. Classification of Herbal Plant and Comparative Analysis of SVM and KNN Classifier Models on the Leaf Features Using Machine Learning // Soft Computing for Intelligent Systems / Ed. by N. Marriwala, C. C. Tripathi, S. Jain, S. Mathapathi. Singapore : Springer, 2021. P. 227–239. DOI: https://doi.org/10.1007/978-981-16-1048-6_17.
Balasubramanian K., Kishore R., Krishnamoorthy G. D. Optimal Knee Osteoarthritis Diagnosis Using Hybrid Deep Belief Network Based on Swarm Optimization Method // Concurrency and Computation: Practice and Experience. 2022. Vol. 34, Iss. 13, Art. No. e6913. DOI: https://doi.org/10.1002/cpe.6913.
Messaoudene K., Harrar K. Computerized Diagnosis of Knee Osteoarthritis from X‐Ray Images Using Combined Texture Features: Data from the Osteoarthritis Initiative // International Journal of Imaging Systems and Technology. 2024. Vol. 34, Iss. 2, Art. No. e23063. DOI: https://doi.org/10.1002/ima.23063.
AI-Assisted Tuberculosis Detection and Classification from Chest X‑Rays Using a Deep Learning Normalization-Free Network Model / V. Acharya, G. Dhiman, K. Prakasha [et al.] // Computational Intelligence and Neuroscience. 2022. Vol. 2022, Art. No. 2399428. DOI: https://doi.org/10.1155/2022/2399428.
This work is licensed under a Creative Commons Attribution-NonCommercial-ShareAlike 4.0 International License
Copyright © 2024 Zhiljakov A. A., Volokitina E. A.