Segmentation of Vertebrae and Diagnosis of Fractures in Magnetic Resonance Imaging Using U-Net 3D and Deep Belief Network

• Anderson Matheus Passos Paiva UFMA
• João Otávio Bandeira Diniz UFMA
• Aristófanes Corrêa Silva UFMA
• Anselmo Cardoso Paiva UFMA

DOI: https://doi.org/10.5753/sbcas.2019.6246

Abstract

Lumbar pain is a common reason for clinical visits and magnetic res- onance imaging is frequently used in systems to support the diagnosis of spinal pathologies. Aiming to improve and automate this process, this study propo- ses the use of computational techniques for the segmentation of vertebrae in magnetic resonance imaging, with the purpose of performing further analysis about pathologies in the spine. To achieve this goal, two Deep Learning archi- tectures are used: U-Net for 3D segmentation and Deep Belief Network for the classification of vertebrae with rupture or not. The results show that U-Net is promising to localize the vertebra region, obtaining an average Dice Coefficient value of 89,51%, thus overcoming several important studies focused on the pro- blem. Classification was also efficient, with values of 94.38% for accuracy and 88.8% for sensitivity.

References

Akkus, Z., Galimzianova, A., Hoogi, A., Rubin, D. L., and Erickson, B. J. (2017). Deep learning for brain mri segmentation: state of the art and future directions. Journal of digital imaging, 30(4):449–459. http://dx.doi.org/10.1007/s10278-017-9983-4

ANA (2019). Spine disorders, deformities and diseases. https://ana-neurosurgery.com/areas-of-expertise/spine-disorders-deformities-diseases/. Acesso em: 15 de fevereiro de 2019.

Brooks, R. A. (1977). A quantitative theory of the hounsfield unit and its application to dual energy scanning. Journal of computer assisted tomography, 1(4):487–493.

Chu, C., Belavỳ, D. L., Armbrecht, G., Bansmann, M., Felsenberg, D., and Zheng, G. (2015). Fully automatic localization and segmentation of 3d vertebral bodies from ct/mr images via a learning based method. PloS one, 10(11):e0143327. http://dx.doi.org/10.1371/journal.pone.0143327

Dice, L. R. (1945). Measures of the amount of ecologic association between species. Ecology, 26(3):297–302. http://dx.doi.org/10.2307/1932409

Egger, J., Nimsky, C., and Chen, X. (2017). Vertebral body segmentation with grow-cut: Initial experience, workflow and practical application. SAGE open medicine, 5:2050312117740984. http://dx.doi.org/10.1177/2050312117740984

Freburger, J. K., Holmes, G. M., Agans, R. P., Jackman, A. M., Darter, J. D., Wallace, A. S., Castel, L. D., Kalsbeek, W. D., and Carey, T. S. (2009). The rising prevalence of chronic low back pain. Archives of internal medicine, 169(3):251–258. http://dx.doi.org/10.1001/archinternmed.2008.543

Hille, G., Saalfeld, S., Serowy, S., and Tönnies, K. (2018). Vertebral body segmentation in wide range clinical routine spine mri data. Computer methods and programs in biomedicine, 155:93–99. http://dx.doi.org/10.1016/j.cmpb.2017.12.013

Juntu, J., Sijbers, J., Van Dyck, D., and Gielen, J. (2005). Bias field correction for mri images. In Computer Recognition Systems, pages 543–551. Springer. http://dx.doi.org/10.1007/3-540-32390-2_64

Korez, R., Likar, B., Pernuš, F., and Vrtovec, T. (2016). Model-based segmentation of vertebral bodies from mr images with 3d cnns. In Ourselin, S., Joskowicz, L., Sabuncu, M. R., Unal, G., and Wells, W., editors, Medical Image Computing and Computer- Assisted Intervention – MICCAI 2016, pages 433–441, Cham. Springer International Publishing. http://dx.doi.org/10.1007/978-3-319-46723-8_50

LeCun, Y. and Cortes, C. (2010). MNIST handwritten digit database.

Lenchik, L., Rogers, L. F., Delmas, P. D., and Genant, H. K. (2004). Diagnosis of osteoporotic vertebral fractures: importance of recognition and description by radiologists. American Journal of Roentgenology, 183(4):949–958. http://dx.doi.org/10.2214/ajr.183.4.1830949

Liu, G., Xiao, L., and Xiong, C. (2017). Image classification with deep belief networks and improved gradient descent. In 2017 IEEE International Conference on Computational Science and Engineering (CSE) and IEEE International Conference on Embedded and Ubiquitous Computing (EUC), volume 1, pages 375–380. http://dx.doi.org/10.1109/CSE-EUC.2017.74

Lu, J.-T., Pedemonte, S., Bizzo, B., Doyle, S., Andriole, K. P., Michalski, M. H., Gonzalez, R. G., and Pomerantz, S. R. (2018). Deepspine: Automated lumbar vertebral segmentation, disc-level designation, and spinal stenosis grading using deep learning. arXiv preprint arXiv:1807.10215.

Mendieta, J. B. (2016). An efficient and semiautomatic segmentation method for 3d surface reconstruction of the lumbar spine from magnetic resonance imaging (mri). Master’s thesis, Queensland University of Technology.

Neubert, A., Fripp, J., Shen, K., Salvado, O., Schwarz, R., Lauer, L., Engstrom, C., and Crozier, S (2011). Automated 3d segmentation of vertebral bodies and intervertebral discs from mri. In 2011 International Conference on Digital Image Computing: Techniques and Applications, pages 19–24. http://dx.doi.org/10.1109/DICTA.2011.12

Richards, P. J., George, J., Metelko, M., and Brown, M. (2010). Spine computed tomography doses and cancer induction. Spine, 35(4):430–433. http://dx.doi.org/10.1097/BRS.0b013e3181cdde47

Ronneberger, O., Fischer, P., and Brox, T. (2015). U-net: Convolutional networks for biomedical image segmentation. In International Conference on Medical image computing and computer-assisted intervention, pages 234–241. Springer. http://dx.doi.org/10.1007/978-3-319-24574-4_28

Schwarzenberg, R., Freisleben, B., Nimsky, C., and Egger, J. (2014). Cube-cut: vertebral body segmentation in mri-data through cubic-shaped divergences. PloS one, 9(4):e93389. http://dx.doi.org/10.1371/journal.pone.0093389

Zou, K. H., Warfield, S. K., Bharatha, A., Tempany, C. M., Kaus, M. R., Haker, S. J., Wells III, W. M., Jolesz, F. A., and Kikinis, R. (2004). Statistical validation of image segmentation quality based on a spatial overlap index1: scientific reports. Academic radiology, 11(2):178–189. http://dx.doi.org/10.1016/S1076-6332(03)00671-8

Zukić, D., Vlasák, A., Egger, J., Hořı́nek, D., Nimsky, C., and Kolb, A. (2014). Robust detection and segmentation for diagnosis of vertebral diseases using routine mr images. In Computer Graphics Forum, volume 33, pages 190–204. Wiley Online Library. http://dx.doi.org/10.1111/cgf.12343