Using artificial neural networks to perform segmentation of hip radiographs in the treatment of osteoarthritis

Today, the X-ray analysis procedure makes it possible to detect osteoarthritis (OA) in the early stages of the disorder. The presence or absence of the disorder is detected only when it has already manifested, and X-ray diagnostics have been carried out. The use of automated procedures for analyzing X-ray images and the availability of archives of such information with a long history can improve the results of predicting complications in patients. The article describes the experience of developing an application for computer analysis of radiographs, which, based on deep learning methods, allows us to identify the risks of developing osteoarthritis of the hip joint. The archive of a specialized medical institute is used as a training sample. In order to increase the size of the training set of radiographs, a data augmentation method is used, which increases the variability of the original data and, in some cases, increases the recognition efficiency. The research uses a convolutional network (U-net) designed for image segmentation, which is trained on X-ray images of a specific medical institution. As part of a project to segment and analyze the geometric characteristics of X-ray images of the hip joints, the software to automate the recognition of the joint space size was developed, which helps to clarify the patient’s diagnosis and prognosis for the development of the pathology.

1. Torshin I.Yu., Lila A.M., Zagorodniy N.V., Nazarenko A.G., Tkacheva O.N., Dudinskaya E.N., Alekseeva L.I., Taskina E.A., Sarvilina I.V., Shavlovskaya O.A., Danilov A.B., Minasov T.B., Galustyan A.N., Maljavskaya S.I., Gromov A.N., Egorova E.Yu., Vasil'eva L.V., Evstratova E.F., Gogoleva I.V., Fedotova L.Ye., Udovika M.I., Maksimov V.A., Povzun A.S., Gromova O.A. Development of a verified osteoarthritis risk scale based on a cross-sectional study of clinical and anamnestic parameters and pharmacological anamnesis of patients. Farmakoekonomika. Sovremennaya farmakoekonomika i farmakojepidemiologiya = Farmakoekonomika. Modern Pharmacoeconomics and Pharmacoepidemiology. 2023;16(1):70–79. (In Russ.).

2. Fedonnikov A.S., Kolesnikova A.S., Rozhkova Yu.Yu., Kossovich L.Yu. Decision making support system in spine-and-pel-vic surgery as an instrument of branch control automation. Saratovskiy nauchno-medicinskiy zhurnal = Saratov Journal of Medical Scientific Research. 2019;15(3):677–682. (In Russ.).

3. Johnson L.G., Bortolussi-Courval S., Chehil A., et al. Application of statistical shape modeling to the human hip joint: a scoping review. JBI Evidence Synthesis. 2023;21(3):533–583. DOI: 10.11124/JBIES-22-00175.

4. Kinds M.B., Marijnissen A.C.A., Vincken K.L., Viergever M.A., Drossaers-Bakker K.W., Bijlsma J.W.J., Bierma-Zeinstra S.M.A., Welsing P.M.J., Lafeber F.P.J.G. Evaluation of separate quantitative radiographic features adds to the prediction of incident radiographic osteoarthritis in individuals with recent onset of knee pain: 5-year follow-up in the CHECK cohort. Osteoarthritis and Cartilage. 2012;20(6):548–556. DOI: doi.org/10.1016/j.joca.2012.02.009.

5. Ivanov D.V., Kirillova I.V., Kossovich L.Yu. Biomechanics as a basis for clinical decision support systems in the surgery of the spine-pelvic complex. In: Advances in Solid and Fracture Mechanics. Advanced structured materials. 2022:99–126. URL: https://pureportal.spbu.ru/files/98221533/editor_HA_et_al_LAST.pdf (accessed on 06.10.2023). DOI: 10.1007/978-3-031-18393-5_7.

6. Homma Y., Baba T., Sumiyoshi Nobuhiko, Ochi H., Kobayashi Hideo, Matsumoto M., Yuasa T., Kaneko K. Rapid Hip Osteoarthritis Development in a Patient with Anterior Acetabular Cyst with Sagittal Alignment Change. Case Report in Orthopedics. 2014. DOI: 10.1155/2014/523426.

7. Daniel M. Mathematical simulation of the hip joint loading. PhD. Dissertation, Czech Technical University in Prague Faculty of Mechanical Engineering, Department of Mechanics, Laboratory of Biomechanics of Man. [Internet] 2004. URL: http://physics.fe.uni-lj.si/members/associate/disertation.pdf (accessed on 15.08.2023).

8. Gong Z., Fu Y., He M., et al. Automated identification of hip arthroplasty implants using artificial intelligence. Scientific Reports. 2022;12. DOI: 10.1038/s41598-022-16534-3.

9. Filist S.A., Kondrashov D.S., Sukhomlinov A.Y., Shulga L.V., Al-Darraji Ch.H., Belozerov V.A. Automated system for classifying pancreatic ultrasound images based on the segment-by-segment spectral analysis method. Modelirovanie, optimizatsiya i informatsionnye tekhnologii = Modeling, Optimization and Information Technology. 2023;11(1). URL: https://moitvivt.ru/ru/journal/pdf?id=1302. DOI: 10.26102/2310-6018/2023.40.1.021 (accessed on 15.10.2023) (In Russ.).

10. Mustra M., Delac K., Grgic M. Overview of the DICOM standard. In: ELMAR, 2008. 50th International Symposium, 10-12 September 2008, Zadar, Croatia. IEEE; 2008. p. 39–44.

11. Lakhani P., Gray D.L., Pett C.R., et al. Hello world deep learning in medical imaging. Journal of Digital Imaging. 2018;31(3):283–289. DOI: 10.1007/s10278-018-0079-6.

12. Russell B.C., Torralba A., Murphy K.P., Freeman W.T. LabelMe: A database and web-based tool for image annotation. International Journal of Computer Vision. 2008;77(1):157–173. DOI: 10.1007/s11263-007-0090-8.

13. Ronneberger O., Fischer Ph., Brox T. U-Net: Convolutional Networks for Biomedical Image Segmentation. In: Medical Image Computing and Computer-Assisted Intervention – MICCAI 2015: 18th International Conference, 5-9 October 2015, Münich, Germany. Springer; 2015. p. 234–241.

14. Siddique N., Paheding S., Elkin C.P., Devabhaktuni V. U-Net and its variants for medical image segmentation: a review of theory and applications. IEEE Access. 2021;9:82031–82057. DOI: 10.1109/ACCESS.2021.3086020.