Advertisement

Fully automated radiological analysis of spinal disorders and deformities: a deep learning approach

  • Fabio GalbuseraEmail author
  • Frank Niemeyer
  • Hans-Joachim Wilke
  • Tito Bassani
  • Gloria Casaroli
  • Carla Anania
  • Francesco Costa
  • Marco Brayda-Bruno
  • Luca Maria Sconfienza
Original Article

Abstract

Purpose

We present an automated method for extracting anatomical parameters from biplanar radiographs of the spine, which is able to deal with a wide scenario of conditions, including sagittal and coronal deformities, degenerative phenomena as well as images acquired with different fields of view.

Methods

The location of 78 landmarks (end plate centers, hip joint centers, and margins of the S1 end plate) was extracted from three-dimensional reconstructions of 493 spines of patients suffering from various disorders, including adolescent idiopathic scoliosis, adult deformities, and spinal stenosis. A fully convolutional neural network featuring an additional differentiable spatial to numerical (DSNT) layer was trained to predict the location of each landmark. The values of some parameters (T4–T12 kyphosis, L1–L5 lordosis, Cobb angle of scoliosis, pelvic incidence, sacral slope, and pelvic tilt) were then calculated based on the landmarks’ locations. A quantitative comparison between the predicted parameters and the ground truth was performed on a set of 50 patients.

Results

The spine shape predicted by the models was perceptually convincing in all cases. All predicted parameters were strongly correlated with the ground truth. However, the standard errors of the estimated parameters ranged from 2.7° (for the pelvic tilt) to 11.5° (for the L1–L5 lordosis).

Conclusions

The proposed method is able to automatically determine the spine shape in biplanar radiographs and calculate anatomical and posture parameters in a wide scenario of clinical conditions with a very good visual performance, despite limitations highlighted by the statistical analysis of the results.

Graphical abstract

These slides can be retrieved under Electronic Supplementary Material.

Keywords

Deep learning Spine deformities Automated analysis Coordinate regression Biplanar radiographs 

Notes

Acknowledgements

The work has been partially funded by the Italian Ministry of Health (Ricerca Corrente). We gratefully acknowledge the support of NVIDIA Corporation with the donation of the Titan Xp GPU used for this research.

Compliance with ethical standards

Conflict of interest

The authors declare that there is no conflict of interest regarding the publication of this article.

Supplementary material

586_2019_5944_MOESM1_ESM.pptx (793 kb)
Supplementary material 1 (PPTX 792 kb)

References

  1. 1.
    Duval-Beaupere G, Schmidt C, Cosson P (1992) A Barycentremetric study of the sagittal shape of spine and pelvis: the conditions required for an economic standing position. Ann Biomed Eng 20:451–462CrossRefGoogle Scholar
  2. 2.
    Le Huec JC, Charosky S, Barrey C, Rigal J, Aunoble S (2011) Sagittal imbalance cascade for simple degenerative spine and consequences: algorithm of decision for appropriate treatment. Eur Spine J 20(Suppl 5):699–703CrossRefGoogle Scholar
  3. 3.
    Le Huec JC, Roussouly P (2011) Sagittal spino-pelvic balance is a crucial analysis for normal and degenerative spine. Eur Spine J 20(Suppl 5):556–557CrossRefGoogle Scholar
  4. 4.
    Ferguson AB (1930) The study and treatment of scoliosis. South Med J 23:116–120CrossRefGoogle Scholar
  5. 5.
    Cobb J (1948) Outline for the study of scoliosis. Instr Course Lect AAOS 5:261–275Google Scholar
  6. 6.
    Carman DL, Browne RH, Birch JG (1990) Measurement of scoliosis and kyphosis radiographs. Intraobserver and interobserver variation. J Bone Joint Surg Am 72:328–333CrossRefGoogle Scholar
  7. 7.
    Vrtovec T, Pernuš F, Likar B (2009) A review of methods for quantitative evaluation of spinal curvature. Eur Spine J 18:593–607CrossRefGoogle Scholar
  8. 8.
    Wu H, Bailey C, Rasoulinejad P, Li S (2018) Automated comprehensive Adolescent Idiopathic Scoliosis assessment using MVC-Net. Med Image Anal 48:1–11CrossRefGoogle Scholar
  9. 9.
    Sun H, Zhen X, Bailey C, Rasoulinejad P, Yin Y, Li S (2017) Direct estimation of spinal Cobb angles by structured multi-output regression. In: Niethammer M et al (ed) Information processing in medical imaging. IPMI 2017. Lecture notes in computer science, vol 10265. Springer, Cham, pp 529–540Google Scholar
  10. 10.
    Wu H, Bailey C, Rasoulinejad P, Li S (2017) Automatic landmark estimation for adolescent idiopathic scoliosis assessment using BoostNet. In: Medical image computing and computer assisted intervention—MICCAI 2017, Quebec City, pp 127–135Google Scholar
  11. 11.
    Zhang J, Lou E, Le LH, Hill DL, Raso JV, Wang Y (2009) Automatic Cobb measurement of scoliosis based on fuzzy Hough transform with vertebral shape prior. J Digital Imaging 22:463CrossRefGoogle Scholar
  12. 12.
    Nibali A, He Z, Morgan S, Prendergast L (2018) Numerical coordinate regression with convolutional neural networks. arXiv preprint arXiv:1801.07372
  13. 13.
    Chollet F (2015) Keras: deep learning for humans. https://github.com/keras-team/keras. Accessed 26 Sept 2018
  14. 14.
    Abadi M, Agarwal A, Barham P, Brevdo E, Chen Z, Citro C, et al. (2015) TensorFlow: large-scale machine learning on heterogeneous systems. https://www.tensorflow.org. Accessed 26 Sept 2018
  15. 15.
    Burkardt J (2012). SPLINE—interpolation and approximation of data. https://people.sc.fsu.edu/~jburkardt/cpp_src/spline/spline.html. Accessed 26 Sept 2018
  16. 16.
    Singer K, Jones T, Breidahl P (1990) A comparison of radiographic and computer-assisted measurements of thoracic and thoracolumbar sagittal curvature. Skelet Radiol 19:21–26CrossRefGoogle Scholar
  17. 17.
    Altman DG, Bland JM (1983) Measurement in medicine: the analysis of method comparison studies. Statistician 32:307–317CrossRefGoogle Scholar
  18. 18.
    Jones E, Oliphant T, Peterson P (2014) SciPy: open source scientific tools for Python. https://www.scipy.org. Accessed 26 Sept 2018
  19. 19.
    Seabold S, Perktold J (2010) Statsmodels: econometric and statistical modeling with python. In: Proceedings of the 9th Python in Science Conference, vol 57, p 61Google Scholar
  20. 20.
    Galbusera F, Bassani T, Costa F, Brayda-Bruno M, Zerbi A, Wilke HJ (2018) Artificial neural networks for the recognition of vertebral landmarks in the lumbar spine. Comput Methods Biomech Biomed Eng Imaging Vis 6(4):447–452CrossRefGoogle Scholar
  21. 21.
    Anitha H, Prabhu G (2012) Automatic quantification of spinal curvature in scoliotic radiograph using image processing. J Med Syst 36:1943–1951CrossRefGoogle Scholar
  22. 22.
    Sardjono TA, Wilkinson MH, Veldhuizen AG, van Ooijen PM, Purnama KE, Verkerke GJ (2013) Automatic Cobb angle determination from radiographic images. Spine (Phila Pa 1976) 38:E1256–E1262CrossRefGoogle Scholar
  23. 23.
    Zhang J, Lou E, Hill DL, Raso JV, Wang Y, Le LH, Shi X (2010) Computer-aided assessment of scoliosis on posteroanterior radiographs. Med Biol Eng Comput 48:185–195CrossRefGoogle Scholar
  24. 24.
    Harrison DE, Harrison DD, Cailliet R, Janik TJ, Holland B (2001) Radiographic analysis of lumbar lordosis: centroid, Cobb, TRALL, and Harrison posterior tangent methods. Spine 26:e235–e242CrossRefGoogle Scholar
  25. 25.
    Briggs AM, Van Dieën JH, Wrigley TV, Greig AM, Phillips B, Lo SK, Bennell KL (2007) Thoracic kyphosis affects spinal loads and trunk muscle force. Phys Ther 87:595–607CrossRefGoogle Scholar
  26. 26.
    Kyrölä KK, Salme J, Tuija J, Tero I, Eero K, Arja H (2018) Intra-and interrater reliability of sagittal spinopelvic parameters on full-spine radiographs in adults with symptomatic spinal disorders. Neurospine 15:175–181CrossRefGoogle Scholar
  27. 27.
    Somoskeöy S, Tunyogi-Csapó M, Bogyó C, Illés T (2012) Accuracy and reliability of coronal and sagittal spinal curvature data based on patient-specific three-dimensional models created by the EOS 2D/3D imaging system. Spine J 12:1052–1059CrossRefGoogle Scholar
  28. 28.
    Carreau JH, Bastrom T, Petcharaporn M, Schulte C, Marks M, Illés T, Somoskeöy S, Newton PO (2014) Computer-generated, three-dimensional spine model from biplanar radiographs: a validity study in idiopathic scoliosis curves greater than 50 degrees. Spine Deform 2:81–88CrossRefGoogle Scholar

Copyright information

© Springer-Verlag GmbH Germany, part of Springer Nature 2019

Authors and Affiliations

  • Fabio Galbusera
    • 1
    Email author
  • Frank Niemeyer
    • 2
  • Hans-Joachim Wilke
    • 2
  • Tito Bassani
    • 1
  • Gloria Casaroli
    • 1
  • Carla Anania
    • 3
  • Francesco Costa
    • 3
  • Marco Brayda-Bruno
    • 4
  • Luca Maria Sconfienza
    • 5
    • 6
  1. 1.Laboratory of Biological Structures MechanicsIRCCS Istituto Ortopedico GaleazziMilanItaly
  2. 2.Institute of Orthopedic Research and Biomechanics, Center for Trauma Research UlmUlm UniversityUlmGermany
  3. 3.Department of NeurosurgeryHumanitas Research HospitalRozzanoItaly
  4. 4.Department of Spine Surgery IIIIRCCS Istituto Ortopedico GaleazziMilanItaly
  5. 5.Unit of Diagnostic and Interventional RadiologyIRCCS Istituto Ortopedico GaleazziMilanItaly
  6. 6.Department of Biomedical Sciences for HealthUniversità degli Studi di MilanoMilanItaly

Personalised recommendations