Comparison of Deep Learning Models and Optimization Algorithms in the Detection of Scoliosis and Spondylolisthesis from X-Ray Images

Year 2024, Volume: 28 Issue: 2, 438 - 451, 30.04.2024

Harun Güneş Cengiz Hark Abdullah Erhan Akkaya

https://doi.org/10.16984/saufenbilder.1246001

Abstract

The spine is composed of pieces of bone called vertebrae that lie between the skull and the tailbone. Various medical conditions can affect the spine. In this study, two types of degenerative diseases, scoliosis, and spondylolisthesis, were studied. Deep AI architectures have recently enabled further disease diagnosis innovation using medical images. Various traditional and deep learning studies use medical images for disease diagnosis in the literature. This study aims to classify spine X-ray images according to three possible conditions (Normal, Scoliosis, and Spondylolisthesis) and to exploit the potential of these X-ray images to detect possible diseases occurring in the spine. The performance of deep learning models and optimization algorithms used in this process was evaluated. The study uses a data set created and/or analyzed during an existing study. This data set consists of images that belong to three different classes: scoliosis, spondylolisthesis, or x-ray images of normal (i.e. healthy) individuals. A total of 338 spine X-ray images, 188 scoliosis images, 79 spondylolisthesis images, and 71 normal images. Six different deep-learning architectures have been used in the study. These architectures are Alexnet, GoogLeNet, ResNet-18, ResNet-50, ResNet-101, and EfficientNet-bo. While working on these deep architectures, each model has been evaluated using different optimization algorithms. These optimization algorithms are RmsProp, SGDM, and Adam. According to the classification processes, the deep learning model with the highest accuracy value was Alexnet, and the optimization algorithm used with it, Sgdm (99.01%), and the training time lasted 38 seconds. According to the classification processes, the deep learning model with the fastest completion time (30 seconds) was Alexnet and the optimization algorithm used with it was RmsProp. An accuracy rate of 98.02% has been obtained in the training of this model.

Keywords

Deep-learning, Classification, Spine X-ray Images, Scoliosis, Spondylolisthesis

Project Number

2

References

• [1] M. Alafeef, M. Fraiwan, H. Alkhalaf, Z. Audat, “Shannon entropy and fuzzy C-means weighting for AI-based diagnosis of vertebral column diseases” Journal of Ambient Intelligence and Humanized Computing, Springer, 2020, vol. 11, pp. 2557-2566.
• [2] M. R. Konieczny, H. Senyurt, R. Krauspe, “Epidemiology of adolescent idiopathic scoliosis,” Journal of children's orthopaedics, London, England, vol 7, no. 1, pp. 3-9, 2013.
• [3] M. Alafeef, Z. Audat, L. Fraiwan, T. Manasreh, “Using deep transfer learning to detect scoliosis and spondylolisthesis from X-ray images,” Plos one, vol. 17, no. 5, pp. e0267851, 2022.
• [4] J. Yang, K. Zhang, H. Fan, Z. Huang, Y. Xiang, J. Yang, H. Lin, “Development and validation of deep learning algorithms for scoliosis screening using back images, ”Communications biology, vol. 2, no. 1, pp. 390, 2019.
• [5] American Association of Neurological Surgeons, “Scoliosis”, Jan. 22, 2023. [Online]. Available:https://www.aans.org/Patients/Neurosurgical-Conditions-andTreatments/Scoliosis
• [6] S. The American Academy of Orthopaedic Surgeons, “Spondylolysis and Spondylolisthesis”, Jan. 22, 2023. [Online]. Available:https://orthoinfo.aaos.org/en/diseases--conditions/spondylolysis-and-spondylolisthesis
• [7] M. A. Deveci, A. Şenköylü, “Gelişimsel spondilolisteziste bel ağrısı: tanı ve tedavi yaklaşımı, ” TOTBİD Dergisi:14, pp. 282-289.
• [8] J. S. Smith, C. I. Shaffrey, K. M. Fu, J. K. Scheer, S. Bess, V. Lafage, C.P. Ames, “Clinical and radiographic evaluation of the adult spinal deformity patient,” in Neurosurgery Clinics, Elsevier, vol. 24, no. 2, pp. 143-156, 2013.
• [9] S. Bess, O. Boachie-Adjei, D. Burton, M. Cunningham, C. Shaffrey, A. Shelokov, International Spine Study Group, “Pain and disability determine treatment modality for older patients with adult scoliosis, while deformity guides treatment for younger patients,” Spine, vol. 34, no. 20, pp. 2186-2190, 2009.
• [10] J. S. Smith, C. I. Shaffrey, S. D. Glassman, S. H. Berven, F. J. Schwab, C. L. Hamill, Spinal Deformity Study Group, “Risk-benefit assessment of surgery for adult scoliosis: an analysis based on patient age,” Spine, vol. 36, no.10, pp. 817-824, 2019.
• [11] M. Fraiwan, Z. Audat, L. Fraiwan, T. Manasreh, “Using deep transfer learning to detect scoliosis and spondylolisthesis from X-ray images,” Plos one, vol.17, no. 5, pp. e0267851, 2022.
• [12] R. F. Masood, I. A Taj, M. B. Khan, M. A. Qureshi, T. Hassan, “Deep learning based vertebral body segmentation with extraction of spinal measurements and disorder disease classification,” in Biomedical Signal Processing and Control, Elsevier, vol. 71, pp. 103230, 2022.
• [13] M. Tajdari, A. Pawar, H. Li, F. Tajdari, A. Maqsood, E. Cleary, W. K. Liu, “Image-based modeling for adolescent idiopathic scoliosis: mechanistic machine learning analysis and prediction,” Computer methods in applied mechanics and engineering, Elsevier, vol. 374, pp. 113590, 2021.
• [14] M. Fraiwan, Z. Audat, L. Fraiwan, T. Manasreh, “Using deep transfer learning to detect scoliosis and spondylolisthesis from X-ray images,” Plos one, vol. 17, no. 5, pp. e0267851, 2022.
• [15] I. Özkan, E. Ülker, “Derin öğrenme ve görüntü analizinde kullanılan derin öğrenme modelleri,” Gaziosmanpaşa Bilimsel Araştırma Dergisi, vol. 6, no. 3, pp. 85-104, 2017.
• [16] M. Turkoglu, D. Hanbay, A. Sengur, “Multi-model LSTM-based convolutional neural networks for detection of apple diseases and pests,” in Journal of Ambient Intelligence and Humanized Computing, Springer, pp. 1-11, 2019.
• [17] A. Krizhevsky, I. Sutskever, G. E. Hinton, “Imagenet classification with deep convolutional neural networks,” Communications of the ACM, vol. 60, no. 6, pp. 84-90, 2017.
• [18] S. Gökalp, İ. Aydın, “Farklı Derin Sinir Ağı Modellerinin Duygu Tanımadaki Performansların Karşılaştırılması,” Muş Alparslan Üniversitesi Mühendislik Mimarlık Fakültesi Dergisi, vol. 2, no.1, pp. 35-43, 2021.
• [19] K. He, X. Zhang, S. Ren, j. Sun, “Deep residual learning for image recognition” In Proceedings of the IEEE conference on computer vision and pattern recognition, pp. 770-778, 2016.
• [20] B. Anadolu, “Dijital hikâye anlatıcılığı bağlamında yapay zekânın sinemaya etkisi: Sunspring ve It’s No Game filmlerinin analizi,” Erciyes İletişim Dergisi, vol.1, pp. 39-56, 2019.
• [21] M. Tan, Q. Le, “Efficientnet: Rethinking model scaling for convolutional neural networks,” In International conference on machine learning, PMLR, pp. 6105-6114, 2019.
• [22] E. Seyyarer, F. Ayata, T. Uçkan, A. Krci, “Derin öğrenmede kullanilan optimizasyon algoritmalarnin uygulanmasi ve kiyaslanmasi,” Computer Science, vol. 5, no. 2, pp. 90-98, 2020.
• [23] M. F. Akca “Gradient Descent Nedir?” Jan.22,2023[Online]. Available:https://medium.com/deep-learning-turkiye/gradient-descent-nedir-3ec6afcb9900.
• [24] S. Ruder “An overview of gradient descent optimization algorithms” Jan. 22,2023 [Online]. Available:https://ruder.io/optimizing-gradient-descent/
• [25] Matlab [Online] Available:https:www.mathworks.com/products/matlab.html
• [26] A. A. Reshi, I. Ashraf, F. Rustam, H. F. Shahzad, A. Mehmood, G. S. Choi, “Diagnosis of vertebral column pathologies using concatenated resampling with machine learning algorithms,” Peer Journal Computer Science, vol 7, pp. e547, 2021.
• [27] T. Colombo, M. Mangone, F. Agostini, A. Bernetti, M. Paoloni, V. Santilli, L. Palagi, “Supervised and unsupervised learning to classify scoliosis and healthy subjects based on non-invasive rasterstereography analysis,” Plos one, vol.16, no. 12, pp. e0261511, 2021.
• [28] H. Wang, T. Zhang, K. M. C. Cheung, G. K. H. Shea, “Application of deep learning upon spinal radiographs to predict progression in adolescent idiopathic scoliosis at first clinic visit,” EClinicalMedicine, vol 42, pp. 101220, 2021.
• [29] Y. Unal, K. Polat, H. E. Kocer, “Pairwise FCM based feature weighting for improved classification of vertebral column disorders,”Computers in biology and medicine,vol. 46, pp. 61-70, 2014.