Skip to main content
SpringerLink
Log in

CNN-based qualitative detection of bone mineral density via diagnostic CT slices for osteoporosis screening

  • Original Article
  • Published:
Osteoporosis International Aims and scope Submit manuscript

Abstract

Summary

The features extracted from diagnostic computed tomography (CT) slices were used to qualitatively detect bone mineral density (BMD) through neural network models, and the evaluation results indicated that it may be a promising approach to perform osteoporosis screening in clinical practice.

Introduction

The purpose of this study is to design a novelty diagnostic method for osteoporosis screening by using the convolutional neural network (CNN), which can be incorporated into the procedure of routine CT diagnostic in medical examination thereby improving the osteoporosis diagnosis and reducing the patient burden.

Methods

The proposed CNN-based method mainly comprises two functional modules to perform qualitative detection of BMD by analyzing the diagnostic 2D CT slice. The first functional module aims to locate and segment the ROI of diagnostic 2D CT slice, called Mark-Segmentation-Network (MS-Net). The second functional module is used to determine the category of BMD by the features of ROI, called BMD-Classification-Network (BMDC-Net). The diagnostic 2D CT slice of pedicle level in lumbar vertebrae (L1) was selected from 3D CT image in our experiments firstly. Then, the trained MS-Net can get the mark image of input original 2D CT slice, thereby obtain the segmentation image. Finally, the trained BMDC-Net can obtain the probability value of normal bone mass, low bone mass, and osteoporosis by inputting the segmentation image. On the basis of network results, the radiologists can provide preliminary qualitative diagnosis results of BMD.

Results

Training of the network was performed on diagnostic 2D CT slices of 150 patients. The network was tested on 63 patients. Each patient corresponds to a 2D CT slice. The proposed MS-Net has an excellent segmentation precision on the shape preservation of different lumbar vertebra. The dice index (DI), pixel accuracy (PA), and intersection over union (IOU) of segmentation results are greater than 0.8. The proposed BMDC-Net achieved an accuracy of 76.65% and an area under the receiver operating characteristic curve of 0.9167.

Conclusions

This study proposed a novel method for qualitative detection of BMD via diagnostic CT slices and it has great potential in clinical applications for osteoporosis screening. The method can potentially reduce the manual burden to radiologists and diagnostic cost to patients.

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

Similar content being viewed by others

References

  1. (2001) NIH consensus development panel on osteoporosis prevention diagnosis, and therapy. Osteoporosis prevention, diagnosis, and therapy. JAMA 285:785–795

  2. Cummings SR, Melton LJ (2002) Epidemiology and outcomes of osteoporotic fractures. Lancet 359:1761–1767

    Article  Google Scholar 

  3. Melton LJ, Cooper C (2001) Magnitude and impact of osteoporosis and fractures. In: Marcus R, Feldman D, Kelsey J (eds) Osteoporosis, 2nd edn. Academic Press, San Diego, pp 557–567

    Chapter  Google Scholar 

  4. Lindsay R (1992) The growing problem of osteoporosis[J]. Osteoporos Int 2(6):267–268

    Article  CAS  Google Scholar 

  5. Sambrook PN (2006) Osteoporosis.[J]. Lancet 367(9527):2010–2018

    Article  CAS  Google Scholar 

  6. Unnanuntana A, Gladnick BP, Donnelly E, Lane JM (2010) The assessment of fracture risk. J Bone Joint Surg (Am Vol) 92(3):743–753

    Article  Google Scholar 

  7. Vestergaard P, Rejnmark L, Mosekilde L (2005) Osteoporosis is markedly underdiagnosed: a nationwide study from Denmark. Osteoporos Int 16(2):134–141

    Article  Google Scholar 

  8. (1994) World Health Organization. Assessment of fracture risk and its application to screening for postmenopausal osteoporosis: report of WHO study group. (Technical report series 843)

  9. Stone KL, Seeley DG, Lui L-y et al (2003) BMD at multiple sites and risk of fracture of multiple types: long-term results from the study of osteoporotic fractures. J Bone Miner Res 18(11):1947–1954

    Article  Google Scholar 

  10. Cummings SR, Bates D, Black DM (2002) Clinical use of bone densitometry: scientific review. JAMA 288(15):1889–1897

    Article  Google Scholar 

  11. Lu Y, Genant HK, Shepherd J, Zhao S, Mathur A, Fuerst TP, Cummings SR (2001) Classification of osteoporosis based on bone mineral densities. J Bone Miner Res 16(5):901–910

    Article  CAS  Google Scholar 

  12. Albanese CV, Diessel E, Genant HK (2003) Clinical applications of body composition measurements using DXA. J Clin Densitom 6(2):75–85

    Article  Google Scholar 

  13. Adams JE (2009) Quantitative computed tomography. Eur J Radiol 71(3):415–424

    Article  Google Scholar 

  14. Shaanthana S, Soelaiman IN, Kok-Yong C (2018) Performance of osteoporosis self-assessment tool (OST) in predicting osteoporosis—a review. Int J Environ Res Public Health 15(7):1445

    Article  Google Scholar 

  15. Li X, Na L, Xiaoguang C (2014) Update on the clinical application of quantitative computed tomography (QCT) in osteoporosis. Curr Radiol Rep 2(10):1–5

    Article  Google Scholar 

  16. Pickhardt PJ, Pooler BD, Lauder T, del Rio AM, Bruce RJ, Binkley N (2013) Opportunistic screening for osteoporosis using abdominal computed tomography scans obtained for other indications. Ann Intern Med 158(8):588–595

    Article  Google Scholar 

  17. Romme EA, Murchison JT, Phang KF et al (2012) Bone attenuation on routine chest CT correlates with bone mineral density on DXA in patients with COPD. Bone Miner Res 27(11):2338–2343

    Article  Google Scholar 

  18. Pan Y, Shi D, Wang H, Chen T, Cui D, Cheng X, Lu Y (2020) Automatic opportunistic osteoporosis screening using low-dose chest computed tomography scans obtained for lung cancer screening. Eur Radiol 30(7):4107–4116

    Article  Google Scholar 

  19. Wang SH, Sun J, Phillips P, Zhao G, Zhang YD (2018) Polarimetric synthetic aperture radar image segmentation by convolutional neural network using graphical processing units. J Real-Time Image Proc 15(3):631–642

    Article  Google Scholar 

  20. Chen L, Bentley P, Mori K et al (2018) DRINet for medical image segmentation. IEEE Trans Med Imaging (99):1–1

  21. Long J, Shelhamer E, Darrell T (2014) Fully convolutional networks for semantic segmentation. IEEE Trans Pattern Anal Mach Intell 39(4):640–651

    Google Scholar 

  22. Ronneberger O, Fischer P, Brox T (2015) U-net: convolutional networks for biomedical image segmentation. Med Image Comput Comput-Assist Interv, Springer, Cham 9351:234–241

    Google Scholar 

  23. Krizhevsky A, Sutskever I, Hinton G (2012) ImageNet classification with deep convolutional neural networks. NIPS. Curran Associates Inc., pp 1097–1105

  24. Simonyan K, Zisserman A (2014) Very deep convolutional networks for large-scale image recognition. arXiv preprint arXiv:1409.1556

  25. Szegedy C, Liu W, Jia Y et al (2015) Going deeper with convolutions. Proc IEEE Conf Comput Vis Pattern Recognit:1–9

  26. He K, Zhang X, Ren S et al (2016) Deep residual learning for image recognition. EEE conference on computer vision and pattern recognition. pp 770–778

  27. Baum T, Bauer JS, Klinder T, Dobritz M, Rummeny EJ, Noël PB, Lorenz C (2014) Automatic detection of osteoporotic vertebral fractures in routine thoracic and abdominal MDCT. Eur Radiol 24(4):872–880

    Article  Google Scholar 

  28. Bar A, Wolf L, Amitai O B et al (2017) Compression fractures detection on CT. SPIE Med Imaging

  29. Tomita N, Cheung YY, Hassanpour S (2018) Deep neural networks for automatic detection of osteoporotic vertebral fractures on CT scans. Comput Biol Med 98:8–15

    Article  Google Scholar 

  30. Zhang M, Gong H, Zhang K (2019) Prediction of lumbar vertebral strength of elderly men based on quantitative computed tomography images using machine learning. Osteoporos Int, 1–12

  31. Huang G, Liu Z, Maaten L V D et al (2017) Densely connected convolutional networks. CVPR, IEEE Comput Soc

  32. Chai T, Draxler RR (2014) Root mean square error (RMSE) or mean absolute error (MAE)? – arguments against avoiding RMSE in the literature. Geosci Model Dev 7(3):1247–1250

    Article  Google Scholar 

  33. Zhou Z, Siddiquee MMR, Tajbakhsh N et al (2018) UNet++: a nested U-Net architecture for medical image segmentation. 11045:3–11

  34. Krizhevsky A, Hinton G (2009) Learning multiple layers of features from tiny images. Tech Report

  35. Wang SH, Lv YD, Sui Y, Liu S, Wang SJ, Zhang YD (2018) Alcoholism detection by data augmentation and convolutional neural network with stochastic pooling. J Med Syst 42(1):2

    Article  Google Scholar 

  36. Y. Jia, E. Shelhamer, J. Donahue et al (2014) Caffe: convolutional architecture for fast feature embedding. ACM Int Conf Multimed 675–678

  37. Kingma DP, Ba J (2015) Adam: A method for stochastic optimization. International Conference on Learning Representations

  38. Srivastava N, Hinton G, Krizhevsky A et al (2014) Dropout: a simple way to prevent neural networks from overfitting. J Mach Learn Res 15(1):1929–1958

    Google Scholar 

  39. Hutenlocher D, Klanderman G, Rucklidge W (1993) Comparing images using the Hausdorff distance. IEEE Trans Pattern Anal Mach Intell 15:850–863

    Article  Google Scholar 

  40. Fawcett T (2005) An introduction to ROC analysis. Pattern Recogn Lett 27(8):861–874

    Article  Google Scholar 

  41. Zysset P, Qin L, Lang T, Khosla S, Leslie WD, Shepherd JA, Schousboe JT, Engelke K (2015) Clinical use of quantitative computed tomography-based finite element analysis of the hip and spine in the management of osteoporosis in adults: the 2015 ISCD official positions-part II. J Clin Densitom 18(3):359–392

    Article  Google Scholar 

  42. Matsumoto T, Ohnishi I, Bessho M, Imai K, Ohashi S, Nakamura K (2009) Prediction of vertebral strength under loading conditions occurring in activities of daily living using a computed tomography-based nonlinear finite element method. Spine (Phila Pa 1976) 34(14):1464–1469

    Article  Google Scholar 

  43. Valentinitsch A, Trebeschi S, Kaesmacher J, Lorenz C, Löffler MT, Zimmer C, Baum T, Kirschke JS (2019) Opportunistic osteoporosis screening in multi-detector CT images via local classification of textures. Osteoporos Int 30(6):1275–1285

    Article  CAS  Google Scholar 

  44. Lee SJ, Pickhardt PJ (2017) Opportunistic screening for osteoporosis using body CT scans obtained for other indications: the UW experience. Clin Rev Bone Mineral Metabol 15:128–137

    Article  CAS  Google Scholar 

Download references

Acknowledgments

We are grateful to all of the participants in this study as well as all of the technicians who performed the scans.

Funding

This work was supported by the Nation Key Research and Development Program of China under grant 2018YFC0114500.

Author information

Authors and Affiliations

  1. PLA Strategy Support Force Information Engineering University, No.62 Science Avenue, Zhengzhou, Henan Province, China

    C. Tang, W. Zhang, L. Li, Z. Li, A. Cai, L. Wang & B. Yan

  2. Department of Radiology in Henan Provincial People’s Hospital, Zhengzhou, Henan Province, China

    H. Li & D. Shi

Authors
  1. C. Tang
    View author publications

    You can also search for this author in PubMed Google Scholar

  2. W. Zhang
    View author publications

    You can also search for this author in PubMed Google Scholar

  3. H. Li
    View author publications

    You can also search for this author in PubMed Google Scholar

  4. L. Li
    View author publications

    You can also search for this author in PubMed Google Scholar

  5. Z. Li
    View author publications

    You can also search for this author in PubMed Google Scholar

  6. A. Cai
    View author publications

    You can also search for this author in PubMed Google Scholar

  7. L. Wang
    View author publications

    You can also search for this author in PubMed Google Scholar

  8. D. Shi
    View author publications

    You can also search for this author in PubMed Google Scholar

  9. B. Yan
    View author publications

    You can also search for this author in PubMed Google Scholar

Corresponding author

Correspondence to B. Yan.

Ethics declarations

Conflicts of interest

None.

Ethical approval

All procedures performed in studies involving human participants were in accordance with the ethical standards of the institutional and/or national research committee and with the 1964 Helsinki declaration and its later amendments or comparable ethical standards.

Informed consent

Informed consent was obtained from all individual participants included in the study.

Additional information

Publisher’s note

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

Supplementary Information

ESM 1

(PPTX 183 kb)

ESM 2

(PPTX 36 kb)

ESM 3

(DOCX 39.7 kb)

Rights and permissions

Reprints and permissions

About this article

Check for updates. Verify currency and authenticity via CrossMark

Cite this article

Tang, C., Zhang, W., Li, H. et al. CNN-based qualitative detection of bone mineral density via diagnostic CT slices for osteoporosis screening. Osteoporos Int 32, 971–979 (2021). https://doi.org/10.1007/s00198-020-05673-w

Download citation

  • Received:

  • Accepted:

  • Published:

  • Issue Date:

  • DOI: https://doi.org/10.1007/s00198-020-05673-w

Keywords

Search

Navigation