Diagnostic triage in patients with central lumbar spinal stenosis using a deep learning system of radiographs

Tackeun Kim MD, MSc1, Young-Gon Kim PhD2,3, Seyeon Park BS2, Jae-Koo Lee MD1, Chang-Hyun Lee MD, PhD1,4, Seung-Jae Hyun MD, PhD1,5, Chi Heon Kim MD, PhD4,5, Ki-Jeong Kim MD, PhD1,5, and Chun Kee Chung MD, PhD4,5,6
View More View Less
  • 1 Department of Neurosurgery, Seoul National University Bundang Hospital, Seongnam;
  • | 2 Transdisciplinary Department of Medicine & Advanced Technology, Seoul National University Hospital, Seoul;
  • | 3 AI Institute, Seoul National University, Seoul;
  • | 4 Department of Neurosurgery, Seoul National University Hospital, Seoul;
  • | 5 Seoul National University College of Medicine, Seoul; and
  • | 6 Department of Brain and Cognitive Sciences, Seoul National University College of Natural Sciences, Seoul, Republic of Korea
Restricted access

Purchase Now

USD  $45.00

Spine - 1 year subscription bundle (Individuals Only)

USD  $376.00

JNS + Pediatrics + Spine - 1 year subscription bundle (Individuals Only)

USD  $612.00
USD  $45.00
USD  $376.00
USD  $612.00
Print or Print + Online Sign in

OBJECTIVE

Magnetic resonance imaging (MRI) is the gold-standard tool for diagnosing lumbar spinal stenosis (LSS), but it is difficult to promptly examine all suspected cases with MRI considering the modality’s high cost and limited accessibility. Although radiography is an efficient screening technique owing to its low cost, rapid operability, and wide availability, its diagnostic accuracy is relatively poor. In this study, the authors aimed to develop a deep learning model with a convolutional neural network (CNN) for diagnosing severe central LSS using radiography and to evaluate radiological diagnostic features using gradient-weighted class activation mapping (Grad-CAM).

METHODS

Patients who had undergone both spinal MRI and radiography in the period from May 1, 2005, to December 31, 2017, were screened. According to the formal MRI report, participants were consecutively included in the severe central LSS or healthy control group, and radiographs for both groups were collected. A CNN-based transfer learning algorithm was developed to classify radiographic findings as LSS or normal (binary classification). The proposed models were evaluated using six performance metrics: area under the receiver operating characteristic curve (AUROC), accuracy, sensitivity, specificity, and positive and negative predictive values.

RESULTS

The VGG19 model achieved the highest accuracy with an AUROC of 90.0% (95% CI 89.8%–90.3%) by training 12,442 images. Accuracy was 82.8% (95% CI 82.5%–83.1%) by averaging 5-fold models. Feature points on Grad-CAM were reasonable, and the features could be categorized into reduced disc height, narrow foramina, short pedicle, and hyperdense facet joint. The AUROC in the extra validation was 89.3% (95% CI 88.7%–90.0%). Accuracy was 81.8% (95% CI 80.6%–83.0%) by averaging 5-fold models. Multivariate logistic regression analysis showed that a combination of demographic factors (age and sex) did not improve the model performance.

CONCLUSIONS

The algorithm trained by a CNN to identify central LSS on radiographs showed high diagnostic accuracy and is expected to be useful as a triage tool. The algorithm could accurately localize the stenotic lesion to assist physicians in the identification of LSS.

ABBREVIATIONS

AI = artificial intelligence; AUROC = area under the receiver operating characteristic curve; CNN = convolutional neural network; CPU = central processing unit; GPU = graphics processing unit; Grad-CAM = gradient-weighted class activation mapping; LSS = lumbar spinal stenosis; MRI = magnetic resonance imaging.

Illustration from Kong et al. (pp 4–12). Copyright Qing-Jie Kong. Used with permission.

Spine - 1 year subscription bundle (Individuals Only)

USD  $376.00

JNS + Pediatrics + Spine - 1 year subscription bundle (Individuals Only)

USD  $612.00
USD  $376.00
USD  $612.00
  • 1

    Huber FA, Stutz S, Vittoria de Martini I, et al. Qualitative versus quantitative lumbar spinal stenosis grading by machine learning supported texture analysis-Experience from the LSOS study cohort. Eur J Radiol. 2019;114:4550.

    • Crossref
    • PubMed
    • Search Google Scholar
    • Export Citation
  • 2

    Lee CH, Chung CK, Kim CH, Kwon JW. Health care burden of spinal diseases in the Republic of Korea: analysis of a nationwide database from 2012 through 2016. Neurospine. 2018;15(1):6676.

    • Crossref
    • PubMed
    • Search Google Scholar
    • Export Citation
  • 3

    Saifuddin A. The imaging of lumbar spinal stenosis. Clin Radiol. 2000;55(8):581594.

  • 4

    Lamb CR, Nicholls A, Targett M, Mannion P. Accuracy of survey radiographic diagnosis of intervertebral disc protrusion in dogs. Vet Radiol Ultrasound. 2002;43(3):222228.

    • Crossref
    • PubMed
    • Search Google Scholar
    • Export Citation
  • 5

    Aggarwal R, Sounderajah V, Martin G, et al. Diagnostic accuracy of deep learning in medical imaging: a systematic review and meta-analysis. NPJ Digit Med. 2021;4(1):65.

    • Crossref
    • PubMed
    • Search Google Scholar
    • Export Citation
  • 6

    Nam KH, Kim DH, Choi BK, Han IH. Internet of things, digital biomarker, and artificial intelligence in spine: current and future perspectives. Neurospine. 2019;16(4):705711.

    • Crossref
    • PubMed
    • Search Google Scholar
    • Export Citation
  • 7

    Lee GY, Lee JW, Choi HS, Oh KJ, Kang HS. A new grading system of lumbar central canal stenosis on MRI: an easy and reliable method. Skeletal Radiol. 2011;40(8):10331039.

    • Crossref
    • PubMed
    • Search Google Scholar
    • Export Citation
  • 8

    He K, Zhang X, Ren S, Sun J. Deep residual learning for image recognition. In: Proceedings of the 29th IEEE International Conference on Computer Vision and Pattern Recognition. IEEE;2016:770778.

    • Search Google Scholar
    • Export Citation
  • 9

    Simonyan K, Zisserman A. Very deep convolutional networks for large-scale image recognition. arXiv. Preprint posted online September 4, 2014.doi: 1409.1556

    • Search Google Scholar
    • Export Citation
  • 10

    Tan M, Le Q. Efficientnet: Rethinking model scaling for convolutional neural networks. In: Proceedings of the 36th International Conference on Machine Learning, PMLR 97. PMLR;2019:61056114.

    • Search Google Scholar
    • Export Citation
  • 11

    Zhou B, Khosla A, Lapedriza A, Oliva A, Torralba A. Learning deep features for discriminative localization. In: Proceedings of the 29th IEEE International Conference on Computer Vision and Pattern Recognition. IEEE;2016:29212929.

    • Search Google Scholar
    • Export Citation
  • 12

    Selvaraju RR, Cogswell M, Das A, Vedantam R, Parikh D, Batra D. Grad-CAM: Visual explanations from deep networks via gradient-based localization. In: Proceedings of the 2017 IEEE International Conference on Computer Vision. IEEE;2017:618626.

    • Search Google Scholar
    • Export Citation
  • 13

    Kim T, Heo J, Jang DK, et al. Machine learning for detecting moyamoya disease in plain skull radiography using a convolutional neural network. EBioMedicine. 2019;40:636642.

    • Crossref
    • PubMed
    • Search Google Scholar
    • Export Citation
  • 14

    Yi PH, Kim TK, Yu AC, Bennett B, Eng J, Lin CT. Can AI outperform a junior resident? Comparison of deep neural network to first-year radiology residents for identification of pneumothorax. Emerg Radiol. 2020;27(4):367375.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 15

    Pham TD. Classification of COVID-19 chest X-rays with deep learning: new models or fine tuning?. Health Inf Sci Syst. 2020;9(1):2.

  • 16

    Üreten K, Erbay H, Maraş HH. Detection of rheumatoid arthritis from hand radiographs using a convolutional neural network. Clin Rheumatol. 2020;39(4):969974.

    • Crossref
    • PubMed
    • Search Google Scholar
    • Export Citation
  • 17

    Joshi RS, Lau D, Ames CP. Artificial intelligence and the future of spine surgery. Neurospine. 2019;16(4):637639.

  • 18

    Perez-Breva L, Shin JH. Artificial intelligence in neurosurgery: a comment on the possibilities. Neurospine. 2019;16(4):640642.

  • 19

    Schwartz JT, Gao M, Geng EA, Mody KS, Mikhail CM, Cho SK. Applications of machine learning using electronic medical records in spine surgery. Neurospine. 2019;16(4):643653.

    • Crossref
    • PubMed
    • Search Google Scholar
    • Export Citation
  • 20

    Kim M, Yun J, Cho Y, et al. Deep learning in medical imaging. Neurospine. 2019;16(4):657668.

  • 21

    Kim S, Bae WC, Masuda K, Chung CB, Hwang D. Fine-grain segmentation of the intervertebral discs from MR spine images using deep convolutional neural networks: BSU-Net. Appl Sci (Basel). 2018;8(9):1656.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 22

    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.

    • Crossref
    • PubMed
    • Search Google Scholar
    • Export Citation
  • 23

    Weng CH, Wang CL, Huang YJ, et al. Artificial intelligence for automatic measurement of sagittal vertical axis using ResUNet framework. J Clin Med. 2019;8(11):E1826.

  • 24

    Joshi RS, Haddad AF, Lau D, Ames CP. Artificial intelligence for adult spinal deformity. Neurospine. 2019;16(4):686694.

  • 25

    Schwartz JT, Cho BH, Tang P, et al. Deep learning automates measurement of spinopelvic parameters on lateral lumbar radiographs. Spine (Phila Pa 1976). 2021;46(12):E671E678.

    • Search Google Scholar
    • Export Citation
  • 26

    Roberson GH, Llewellyn HJ, Taveras JM. The narrow lumbar spinal canal syndrome. Radiology. 1973;107(1):8997.

Metrics

All Time Past Year Past 30 Days
Abstract Views 1747 1746 233
Full Text Views 250 250 51
PDF Downloads 280 280 52
EPUB Downloads 0 0 0