Skip to main content

Advertisement

SpringerLink
Log in

Automating Scoliosis Measurements in Radiographic Studies with Machine Learning: Comparing Artificial Intelligence and Clinical Reports

  • Original Paper
  • Published:
Journal of Digital Imaging Aims and scope Submit manuscript

Abstract

Scoliosis is a condition of abnormal lateral spinal curvature affecting an estimated 2 to 3% of the US population, or seven million people. The Cobb angle is the standard measurement of spinal curvature in scoliosis but is known to have high interobserver and intraobserver variability. Thus, the objective of this study was to build and validate a system for automatic quantitative evaluation of the Cobb angle and to compare AI generated and human reports in the clinical setting. After IRB was obtained, we retrospectively collected 2150 frontal view scoliosis radiographs at a tertiary referral center (January 1, 2019, to January 1, 2021, ≥ 16 years old, no hardware). The dataset was partitioned into 1505 train (70%), 215 validation (10%), and 430 test images (20%). All thoracic and lumbar vertebral bodies were segmented with bounding boxes, generating approximately 36,550 object annotations that were used to train a Faster R-CNN Resnet-101 object detection model. A controller algorithm was written to localize vertebral centroid coordinates and derive the Cobb properties (angle and endplate) of dominant and secondary curves. AI-derived Cobb angle measurements were compared to the clinical report measurements, and the Spearman rank-order demonstrated significant correlation (0.89, p < 0.001). Mean difference between AI and clinical report angle measurements was 7.34° (95% CI: 5.90–8.78°), which is similar to published literature (up to 10°). We demonstrate the feasibility of an AI system to automate measurement of level-by-level spinal angulation with performance comparable to radiologists.

This is a preview of subscription content, log in via an institution to check access.

Access this article

Log in via an institution

Price excludes VAT (USA)
Tax calculation will be finalised during checkout.

Instant access to the full article PDF.

Institutional subscriptions

Fig. 1
Fig. 2
Fig. 3
Fig. 4
Fig. 5
Fig. 6
Fig. 7
Fig. 8

Similar content being viewed by others

Code Availability

Our end-to-end AI system is part of a freely available web application: https://stanford.edu/~baodo/scoliosisai.htm.

References

  1. National Scoliosis Foundation [August 30, 2020]. NSF is a patient-led nonprofit organization dedicated to helping children, parents, adults, and health-care providers to understand the complexities of spinal deformities such as scoliosis.]. Available from: https://www.scoliosis.org/info.php

  2. Cobb JR. Outline for the study of scoliosis. American Academy of Orthopaedic Surgeons. Instr Course Lect. 1948;5:261-75.

    Google Scholar 

  3. Carman DL, Browne RH, Birch JG. Measurement of scoliosis and kyphosis radiographs. Intraobserver and interobserver variation. J Bone Joint Surg Am. 1990;72(3):328–33. Epub 1990/03/01. PubMed PMID: 2312528.

  4. Loder RT, Urquhart A, Steen H, Graziano G, Hensinger RN, Schlesinger A, et al. Variability in Cobb angle measurements in children with congenital scoliosis. J Bone Joint Surg Br. 1995;77(5):768-70. Epub 1995/09/01. PubMed PMID: 7559707.

    Article  CAS  Google Scholar 

  5. Morrissy RT, Goldsmith GS, Hall EC, Kehl D, Cowie GH. Measurement of the Cobb angle on radiographs of patients who have scoliosis. Evaluation of intrinsic error. J Bone Joint Surg Am. 1990;72(3):320–7. Epub 1990/03/01. PubMed PMID: 2312527.

  6. Pruijs JE, Hageman MA, Keessen W, van der Meer R, van Wieringen JC. Variation in Cobb angle measurements in scoliosis. Skeletal Radiol. 1994;23(7):517–20. Epub 1994/10/01. https://doi.org/10.1007/BF00223081. PubMed PMID: 7824978.

  7. Gstoettner M, Sekyra K, Walochnik N, Winter P, Wachter R, Bach CM. Inter- and intraobserver reliability assessment of the Cobb angle: manual versus digital measurement tools. Eur Spine J. 2007;16(10):1587–92. Epub 2007/06/06. https://doi.org/10.1007/s00586-007-0401-3. PubMed PMID: 17549526; PubMed Central PMCID: PMCPMC2078306.

  8. Zhang J, Lou E, Hill DL, Raso JV, Wang Y, Le LH, et al. Computer-aided assessment of scoliosis on posteroanterior radiographs. Med Biol Eng Comput. 2010;48(2):185–95. Epub 2009/12/17. https://doi.org/10.1007/s11517-009-0556-7. PubMed PMID: 20012376.

  9. Zhang J, Lou E, Le LH, Hill DL, Raso JV, Wang Y. Automatic Cobb measurement of scoliosis based on fuzzy Hough Transform with vertebral shape prior. J Digit Imaging. 2009;22(5):463–72. Epub 2008/06/03. https://doi.org/10.1007/s10278-008-9127-y. PubMed PMID: 18516643; PubMed Central PMCID: PMCPMC3043716.

  10. Zhang J, Lou E, Shi X, Wang Y, Hill DL, Raso JV, et al. A computer-aided Cobb angle measurement method and its reliability. J Spinal Disord Tech. 2010;23(6):383–7. Epub 2010/02/04. https://doi.org/10.1097/BSD.0b013e3181bb9a3c. PubMed PMID: 20124919.

  11. Anitha H, Prabhu GK, Karunakar AK. Reliable and reproducible classification system for scoliotic radiograph using image processing techniques. J Med Syst. 2014;38(11):124. Epub 2014/09/28. https://doi.org/10.1007/s10916-014-0124-z. PubMed PMID: 25261171.

  12. Anitha H, Prabhu GK. Automatic quantification of spinal curvature in scoliotic radiograph using image processing. J Med Syst. 2012;36(3):1943–51. Epub 2011/01/27. https://doi.org/10.1007/s10916-011-9654-9. PubMed PMID: 21267773.

  13. Mukherjee J, Kundu R, Chakrabarti A. Variability of Cobb Angle Measurement from Digital X-ray Image Based on Different De-noising Techniques. International Journal of Biomedical Engineering and Technology, Inderscience. 2014;16. https://doi.org/10.1504/IJBET.2014.065656.

  14. Horng MH, Kuok CP, Fu MJ, Lin CJ, Sun YN. Cobb Angle Measurement of Spine from X-Ray Images Using Convolutional Neural Network. Comput Math Methods Med. 2019;2019:6357171. Epub 2019/04/19. https://doi.org/10.1155/2019/6357171. PubMed PMID: 30996731; PubMed Central PMCID: PMCPMC6399566.

  15. Liu J, Yuan C, Sun X, Sun L, Dong H, Peng Y. The measurement of Cobb angle based on spine X-ray images using multi-scale convolutional neural network. Physical and Engineering Sciences in Medicine. 2021;44(3):809-21. doi: https://doi.org/10.1007/s13246-021-01032-z.

    Article  PubMed  Google Scholar 

  16. Hallinan JTPD, Zhu L, Yang K, Makmur A, Algazwi DAR, Thian YL, et al. Deep Learning Model for Automated Detection and Classification of Central Canal, Lateral Recess, and Neural Foraminal Stenosis at Lumbar Spine MRI. Radiology. 2021:204289. https://doi.org/10.1148/radiol.2021204289.

  17. Sa R, Owens W, Wiegand R, Studin M, Capoferri D, Barooha K, et al., editors. Intervertebral disc detection in X-ray images using faster R-CNN. 2017 39th Annual International Conference of the IEEE Engineering in Medicine and Biology Society (EMBC). 11–15 July 2017.

  18. Abadi Mi, Barham P, Chen J, Chen Z, Davis A, Dean J, et al. TensorFlow: A system for large-scale machine learning. arXiv pre-print server. 2016. arxiv:1605.08695

  19. Ren S, He K, Girshick R, Sun J. Faster R-CNN: Towards Real-Time Object Detection with Region Proposal Networks. IEEE Trans Pattern Anal Mach Intell. 2017;39(6):1137–49. Epub 2016/06/14. https://doi.org/10.1109/tpami.2016.2577031. PubMed PMID: 27295650.

  20. Girshick R. Fast R-CNN. arXiv pre-print server. 2015. arxiv:1504.08083.

  21. Lin T-Y, Maire M, Belongie S, Hays J, Perona P, Ramanan D, et al. editors. Microsoft COCO: Common Objects in Context. Computer Vision – ECCV 2014. Cham: Springer International Publishing.

  22. Qian N. On the momentum term in gradient descent learning algorithms. Neural Netw. 1999;12(1):145–51. Epub 2003/03/29. https://doi.org/10.1016/s0893-6080(98)00116-6. PubMed PMID: 12662723.

  23. Allen B. AI Use Cases with Clinical Promise 2018 [March 6, 2021]. Available from: https://www.acrdsi.org/DSIBlog/2018/10/16/21/55/AI-Use-Cases-with-Clinical-Promise.

  24. Galbusera F, Niemeyer F, Wilke HJ, Bassani T, Casaroli G, Anania C, et al. Fully automated radiological analysis of spinal disorders and deformities: a deep learning approach. Eur Spine J. 2019;28(5):951–60. Epub 2019/03/14. https://doi.org/10.1007/s00586-019-05944-z. PubMed PMID: 30864061.

Download references

Author information

Authors and Affiliations

  1. Department of Radiology, Stanford University, 300 Pasteur Drive, Stanford, CA, 94305, USA

    Audrey Y. Ha, Bao H. Do, Adam L. Bartret, Charles X. Fang, Amelie M. Lutz, Geoffrey M. Riley, Daniel L. Rubin, Kathryn J. Stevens, Erin Wang, Shannon Wang & Christopher F. Beaulieu

  2. Department of Radiology, University of California San Diego, 9300 Campus Point Drive, La Jolla, CA, 92037, USA

    Albert Hsiao & Brian Hurt

  3. Department of Radiology, Mayo Clinic, 5779 E Mayo Blvd, Phoenix, AZ, 85054, USA

    Imon Banerjee

Authors
  1. Audrey Y. Ha
    View author publications

    You can also search for this author in PubMed Google Scholar

  2. Bao H. Do
    View author publications

    You can also search for this author in PubMed Google Scholar

  3. Adam L. Bartret
    View author publications

    You can also search for this author in PubMed Google Scholar

  4. Charles X. Fang
    View author publications

    You can also search for this author in PubMed Google Scholar

  5. Albert Hsiao
    View author publications

    You can also search for this author in PubMed Google Scholar

  6. Amelie M. Lutz
    View author publications

    You can also search for this author in PubMed Google Scholar

  7. Imon Banerjee
    View author publications

    You can also search for this author in PubMed Google Scholar

  8. Geoffrey M. Riley
    View author publications

    You can also search for this author in PubMed Google Scholar

  9. Daniel L. Rubin
    View author publications

    You can also search for this author in PubMed Google Scholar

  10. Kathryn J. Stevens
    View author publications

    You can also search for this author in PubMed Google Scholar

  11. Erin Wang
    View author publications

    You can also search for this author in PubMed Google Scholar

  12. Shannon Wang
    View author publications

    You can also search for this author in PubMed Google Scholar

  13. Christopher F. Beaulieu
    View author publications

    You can also search for this author in PubMed Google Scholar

  14. Brian Hurt
    View author publications

    You can also search for this author in PubMed Google Scholar

Contributions

Study supervision and guarantor of the article: Brian Hurt. Analysis and control of the data: Audrey Y. Ha, Bao H. Do, Adam L. Bartret, Charles X. Fang, Erin Wang, Shannon Wang, Brian Hurt. Data interpretation, review and revision of the manuscript: all authors.

Corresponding author

Correspondence to Brian Hurt.

Ethics declarations

Ethics Approval

We retrospectively obtained all scoliosis radiographs from our tertiary care center under Institutional Review Board approval, and informed consent was waived for this HIPAA compliant imaging review.

Conflict of Interest

There are no conflicts of interests that pertain specifically to this work. However, some of the authors are consultants for the medical industry and received grants not related to this study. Albert Hsiao is a founder and consultant for Arterys Inc. and receives grant support from GE Healthcare, Bayer AG, and the American Roentgen Ray Society as an ARRS Scholar, unrelated to this study. Brian Hurt provides consulting services to Imidex Inc unrelated to this study, and supported by the NIH T32EB005970. Amelie M. Lutz receives research funding from GE Healthcare and material support from Bracco Diagnostics Inc. for projects not related to this study. Kathryn J. Stevens receives research funding from GE Healthcare for projects not related to this study.

Additional information

Publisher's Note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Rights and permissions

Reprints and permissions

About this article

Check for updates. Verify currency and authenticity via CrossMark

Cite this article

Ha, A.Y., Do, B.H., Bartret, A.L. et al. Automating Scoliosis Measurements in Radiographic Studies with Machine Learning: Comparing Artificial Intelligence and Clinical Reports. J Digit Imaging 35, 524–533 (2022). https://doi.org/10.1007/s10278-022-00595-x

Download citation

  • Received:

  • Revised:

  • Accepted:

  • Published:

  • Issue Date:

  • DOI: https://doi.org/10.1007/s10278-022-00595-x

Keywords

Search

Navigation