Skip to main content

Advertisement

SpringerLink
Log in

Application of deep learning neural network in predicting bone mineral density from plain X-ray radiography

  • Original Article
  • Published:
Archives of Osteoporosis Aims and scope Submit manuscript

Abstract

Summary

DeepDXA is a deep learning model designed to infer bone mineral density data from plain pelvis X-ray, and it can achieve good predicted value for clinical use.

Purpose

Osteoporosis is defined as a systemic disease of the bone characterized by a decrease in bone strength and deterioration of bone structure at the microscopic level, leading to bone fragility and increased risk of fracture. Bone mineral density (BMD) is the preferred method for the diagnosis of osteoporosis, and dual-energy x-ray absorptiometry (DXA) is the gold standard for diagnosing osteoporosis. Conventional radiography is more suited for the screening of osteoporosis rather than diagnosis, and osteoporosis can be detected on radiographs by experienced physicians only. This study explored the possibility of predicting BMD relative to DXA using patient radiographs.

Methods

A deep learning algorithm of convolutional neural network (CNN) was used for the purpose. The method includes image segmentation, CNN learning, and a convolution-based regression model (DeepDXA) that links the isolated images of the femur bone to predict BMD value. Data were obtained in a single medical center from 2006 to 2018, with a total amount of 3472 pairs of pelvis X-ray and DXA examination within 1 year.

Results

The proposed workflow successfully predicted BMD values of the femur bone with the correlation coefficient (R) of 0.85 (P < 0.001) and the accuracy of 0.88 for prediction osteoporosis, a finding that could be reliably ready for further clinical use.

Conclusion

When suspicious osteoporosis is seen on plain films using the deep learning method we developed, further referral to DXA for the definite diagnosis of osteoporosis is indicated.

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

Similar content being viewed by others

References

  1. Consensus A (1993) Consensus development conference: diagnosis, prophylaxis, and treatment of osteoporosis. Am J Med 94(6):646–650

    Article  Google Scholar 

  2. Solomon C, Black D, Rosen C (2016) Postmenopausal osteoporosis. N Engl J Med 374(3):254–262

    Article  Google Scholar 

  3. Shao C-J, Hsieh Y-H, Tsai C-H, Lai K-A (2009) A nationwide seven-year trend of hip fractures in the elderly population of Taiwan. Bone 44(1):125–129

    Article  Google Scholar 

  4. Choksi P, Jepsen KJ, Clines GA (2018) The challenges of diagnosing osteoporosis and the limitations of currently available tools. Clinical Diabetes and Endocrinology 4(1):12. https://doi.org/10.1186/s40842-018-0062-7

    Article  PubMed  PubMed Central  Google Scholar 

  5. Keyak JH, Skinner HB, Fleming JA (2001) Effect of force direction on femoral fracture load for two types of loading conditions. J Orthop Res 19(4):539–544

    Article  CAS  Google Scholar 

  6. Prevention O (2000) Diagnosis, and therapy. NIH Consens Statement 17(1):1–36

    Google Scholar 

  7. Kanis JA (1994) Assessment of fracture risk and its application to screening for postmenopausal osteoporosis: synopsis of a WHO report. Osteoporos Int 4(6):368–381

    Article  CAS  Google Scholar 

  8. Kanis JA (2002) Diagnosis of osteoporosis and assessment of fracture risk. The Lancet 359(9321):1929–1936

    Article  Google Scholar 

  9. Kanis JA, Melton LJ III, Christiansen C, Johnston CC, Khaltaev N (1994) The diagnosis of osteoporosis. J Bone Miner Res 9(8):1137–1141

    Article  CAS  Google Scholar 

  10. Baim S, Binkley N, Bilezikian JP, Kendler DL, Hans DB, Lewiecki EM, Silverman S (2008) Official positions of the International Society for Clinical Densitometry and executive summary of the 2007 ISCD Position Development Conference. J Clin Densitom 11(1):75–91

    Article  Google Scholar 

  11. Schousboe JT, Shepherd JA, Bilezikian JP, Baim S (2013) Executive summary of the 2013 International Society for Clinical Densitometry Position Development Conference on bone densitometry. J Clin Densitom 16 (4):455–466. doi:https://doi.org/10.1016/j.jocd.2013.08.004

  12. Anil G, Guglielmi G, Peh WC (2010) Radiology of osteoporosis. Radiol Clin North Am 48(3):497–518. https://doi.org/10.1016/j.rcl.2010.02.016

    Article  PubMed  Google Scholar 

  13. Genant HK, Wu CY, Van Kuijk C, Nevitt MC (1993) Vertebral fracture assessment using a semiquantitative technique. J Bone Miner Res 8(9):1137–1148

    Article  CAS  Google Scholar 

  14. Nguyen BNT, Hoshino H, Togawa D, Matsuyama Y (2018) Cortical thickness index of the proximal femur: a radiographic parameter for preliminary assessment of bone mineral density and osteoporosis status in the age 50 years and over population. Clin Orthop Surg 10(2):149–156

    Article  Google Scholar 

  15. He Q, Sun H, Shu L, Zhu Y, Xie X, Zhan Y, Luo C (2018) Radiographic predictors for bone mineral loss: cortical thickness and index of the distal femur. Bone & joint research 7(7):468–475

    Article  Google Scholar 

  16. Clavert P, Javier R-M, Charrissoux J, Obert L, Pidhorz L, Sirveaux F, Mansat P, Fabre T (2016) How to determine the bone mineral density of the distal humerus with radiographic tools? Surg Radiol Anat 38(4):389–393

    Article  Google Scholar 

  17. Samelson EJ, Broe KE, Xu H, Yang L, Boyd S, Biver E, Szulc P, Adachi J, Amin S, Atkinson E, Berger C, Burt L, Chapurlat R, Chevalley T, Ferrari S, Goltzman D, Hanley DA, Hannan MT, Khosla S, Liu C-T, Lorentzon M, Mellstrom D, Merle B, Nethander M, Rizzoli R, Sornay-Rendu E, Van Rietbergen B, Sundh D, Wong AKO, Ohlsson C, Demissie S, Kiel DP, Bouxsein ML (2019) Cortical and trabecular bone microarchitecture as an independent predictor of incident fracture risk in older women and men in the Bone Microarchitecture International Consortium (BoMIC): a prospective study. Lancet Diabetes Endocrinol 7(1):34–43. https://doi.org/10.1016/S2213-8587(18)30308-5

    Article  PubMed  Google Scholar 

  18. Vasikaran S, Eastell R, Bruyère O, Foldes A, Garnero P, Griesmacher A, McClung M, Morris HA, Silverman S, Trenti T (2011) Markers of bone turnover for the prediction of fracture risk and monitoring of osteoporosis treatment: a need for international reference standards. Osteoporos Int 22(2):391–420

    Article  CAS  Google Scholar 

  19. Johnell O, Odén A, De Laet C, Garnero P, Delmas P, Kanis J (2002) Biochemical indices of bone turnover and the assessment of fracture probability. Osteoporos Int 13(7):523

    Article  CAS  Google Scholar 

  20. Garnero P, Sornay-Rendu E, Duboeuf F, Delmas PD (1999) Markers of bone turnover predict postmenopausal forearm bone loss over 4 years: the OFELY study. J Bone Miner Res 14(9):1614–1621

    Article  CAS  Google Scholar 

  21. Silva BC, Leslie WD, Resch H, Lamy O, Lesnyak O, Binkley N, McCloskey EV, Kanis JA, Bilezikian JP (2014) Trabecular bone score: a noninvasive analytical method based upon the DXA image. J Bone Miner Res 29(3):518–530

    Article  Google Scholar 

  22. Winzenrieth R, Michelet F, Hans D (2013) Three-dimensional (3D) microarchitecture correlations with 2D projection image gray-level variations assessed by trabecular bone score using high-resolution computed tomographic acquisitions: effects of resolution and noise. J Clin Densitom 16(3):287–296

    Article  Google Scholar 

  23. Harvey NC, Glüer CC, Binkley N, McCloskey EV, Brandi ML, Cooper C, Kendler D, Lamy O, Laslop A, Camargos BM, Reginster JY, Rizzoli R, Kanis JA (2015) Trabecular bone score (TBS) as a new complementary approach for osteoporosis evaluation in clinical practice. Bone 78:216–224. https://doi.org/10.1016/j.bone.2015.05.016

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  24. Hans D, Barthe N, Boutroy S, Pothuaud L, Winzenrieth R, Krieg M-A (2011) Correlations between trabecular bone score, measured using anteroposterior dual-energy X-ray absorptiometry acquisition, and 3-dimensional parameters of bone microarchitecture: an experimental study on human cadaver vertebrae. J Clin Densitom 14(3):302–312. https://doi.org/10.1016/j.jocd.2011.05.005

    Article  PubMed  Google Scholar 

  25. Martineau P, Silva BC, Leslie WD (2017) Utility of trabecular bone score in the evaluation of osteoporosis. Curr Opin Endocrinol Diabetes Obes 24(6):402–410. https://doi.org/10.1097/med.0000000000000365

    Article  PubMed  Google Scholar 

  26. Silva BC, Leslie WD (2017) Trabecular bone score: a new DXA–derived measurement for fracture risk assessment. Endocrinol Metab Clin 46(1):153–180

    Article  Google Scholar 

  27. Derkatch S, Kirby C, Kimelman D, Jozani MJ, Davidson JM, Leslie WD (2019) Identification of vertebral fractures by convolutional neural networks to predict nonvertebral and hip fractures: a registry-based cohort study of dual X-ray absorptiometry. Radiology 293(2):405–411

    Article  Google Scholar 

  28. Melamed A, Vittinghoff E, Sriram U, Schwartz AV, Kanaya AM (2010) BMD reference standards among South Asians in the United States. J Clin Densitom 13(4):379–384. https://doi.org/10.1016/j.jocd.2010.05.007

    Article  PubMed  PubMed Central  Google Scholar 

  29. Cundy T, Cornish J, Evans MC, Gamble G, Stapleton J, Reid IR (1995) Sources of interracial variation in bone mineral density. J Bone Miner Res 10(3):368–373

    Article  CAS  Google Scholar 

  30. Lindner C, Thiagarajah S, Wilkinson JM, Wallis GA, Cootes TF, Consortium a (2013) Fully automatic segmentation of the proximal femur using random forest regression voting. IEEE transactions on medical imaging 32 (8):1462-1472

  31. He K, Zhang X, Ren S, Sun J Deep residual learning for image recognition. In: Proceedings of the IEEE conference on computer vision and pattern recognition, 2016. pp 770–778

  32. Simonyan K, Zisserman A (2014) Very deep convolutional networks for large-scale image recognition. arXiv preprint arXiv:14091556

  33. Zhang H, Xue J, Dana K Deep ten: Texture encoding network. In: Proceedings of the IEEE conference on computer vision and pattern recognition, 2017. pp 708–717

  34. Yu M, Tham Y-C, Rim TH, Ting DSW, Wong TY, Cheng C-Y (2019) Reporting on deep learning algorithms in health care. The Lancet Digital Health 1(7):e328–e329. https://doi.org/10.1016/S2589-7500(19)30132-3

    Article  PubMed  Google Scholar 

  35. Tecle N, Teitel J, Morris MR, Sani N, Mitten D, Hammert WC (2020) Convolutional neural network for second metacarpal radiographic osteoporosis screening. The Journal of Hand Surgery. https://doi.org/10.1016/j.jhsa.2019.11.019

    Article  PubMed  Google Scholar 

  36. Yasaka K, Akai H, Kunimatsu A, Kiryu S, Abe O (2020) Prediction of bone mineral density from computed tomography: application of deep learning with a convolutional neural network. Eur Radiol. https://doi.org/10.1007/s00330-020-06677-0

    Article  PubMed  Google Scholar 

  37. Liu J, Wang J, Ruan W, Lin C, Chen D (2019) Diagnostic and gradation model of osteoporosis based on improved deep U-Net network. J Med Syst 44(1):15. https://doi.org/10.1007/s10916-019-1502-3

    Article  PubMed  Google Scholar 

  38. Gay J, Harlin H (2019) Texture-based classification for oral cancer detection: implementation and performance analysis of deep learning approaches. Networks (RotEqNet) 8:11

    Google Scholar 

  39. Hu J, Song W, Zhang W, Zhao Y, Yilmaz A (2019) Deep learning for use in lumber classification tasks. Wood Sci Technol 53(2):505–517

    Article  CAS  Google Scholar 

  40. Riggs BL, Melton LJ, Robb RA, Camp JJ, Atkinson EJ, McDaniel L, Amin S, Rouleau PA, Khosla S (2008) A population-based assessment of rates of bone loss at multiple skeletal sites: evidence for substantial trabecular bone loss in young adult women and men. J Bone Miner Res 23(2):205–214. https://doi.org/10.1359/jbmr.071020

    Article  PubMed  Google Scholar 

  41. Herrera A, Lobo-Escolar A, Mateo J, Gil J, Ibarz E, Gracia L (2012) Male osteoporosis: a review. World J Orthop 3(12):223–234. https://doi.org/10.5312/wjo.v3.i12.223

    Article  PubMed  PubMed Central  Google Scholar 

  42. Wang L, Ran L, Zha X, Zhao K, Yang Y, Shuang Q, Liu Y, Hind K, Cheng X, Blake GM (2020) Adjustment of DXA BMD measurements for anthropometric factors and its impact on the diagnosis of osteoporosis. Arch Osteoporos 15(1):155. https://doi.org/10.1007/s11657-020-00833-1

    Article  PubMed  Google Scholar 

  43. Lochmüller EM, Miller P, Bürklein D, Wehr U, Rambeck W, Eckstein F (2000) In situ femoral dual-energy X-ray absorptiometry related to ash weight, bone size and density, and its relationship with mechanical failure loads of the proximal femur. Osteoporos Int 11(4):361–367. https://doi.org/10.1007/s001980070126

    Article  PubMed  Google Scholar 

  44. Boskey AL, Imbert L (2017) Bone quality changes associated with aging and disease: a review. Ann N Y Acad Sci 1410(1):93–106. https://doi.org/10.1111/nyas.13572

    Article  PubMed  PubMed Central  Google Scholar 

  45. Fox KM, Magaziner J, Hawkes WG, Yu-Yahiro J, Hebel JR, Zimmerman SI, Holder L, Michael R (2000) Loss of bone density and lean body mass after hip fracture. Osteoporos Int 11(1):31–35. https://doi.org/10.1007/s001980050003

    Article  CAS  PubMed  Google Scholar 

Download references

Acknowledgements

The authors thank the statistical assistance and wish to acknowledge the support of the Maintenance Project of the Center for Artificial Intelligence in Medicine at Chang Gung Memorial Hospital (Grant CLRPG3H0012, CIRPG3H0012) for study design and monitor, data analysis, and interpretation and Chang Gung Medical Foundation Grant (CMRPG5H0051-3, CMRPG3K0231-2) for manpower and data analysis.

Author information

Author notes

    Authors and Affiliations

    1. Department of Physical Medicine and Rehabilitation, Chang Gung Memorial Hospital at Linkou, No. 5, Fuxing St., Guishan Dist., Taoyuan City, 333, Taiwan

      Chan-Shien Ho & Yu-Cheng Pei

    2. Center for Artificial Intelligence in Medicine, Chang Gung Memorial Hospital at Linkou, No. 5, Fuxing St., Guishan Dist., Taoyuan City, 333, Taiwan

      Yueh-Peng Chen, Tzuo-Yau Fan & Chang-Fu Kuo

    3. Department of Industrial Design, College of Management, Chang Gung University, Taoyuan, Taiwan

      Yueh-Peng Chen

    4. Division of Rheumatology, Allergy and Immunology, Chang Gung Memorial Hospital at Linkou, No. 5, Fuxing St., Guishan Dist., Taoyuan City, 333, Taiwan

      Chang-Fu Kuo

    5. Division of Rheumatology, Orthopaedics and Dermatology, School of Medicine, University of Nottingham, Nottingham, NG7 2UH, UK

      Chang-Fu Kuo

    6. School of Medicine, Chang Gung University, No. 259, Wenhua 1st Rd., Guishan Dist., Taoyuan City, 33302, Taiwan

      Tzu-Yun Yen & Yu-Cheng Pei

    7. Department of Education, Chang Gung Memorial Hospital at Linkou, No.5, Fuxing St., Guishan Dist., Taoyuan City, 333, Taiwan

      Tzu-Yun Yen

    8. Department of Medical Imaging and Intervention, Chang Gung Memorial Hospital, Linkou, No. 5, Fuxing St., Guishan Dist., Taoyuan City, 333, Taiwan

      Yuan-Chang Liu

    9. College of Medicine, Institute for Radiologic Research, Chang Gung University, No. 259, Wenhua 1st Rd., Guishan Dist., Taoyuan City, 33302, Taiwan

      Yuan-Chang Liu

    10. Center of Vascularized Tissue Allograft, Gung Memorial Hospital at Linkou, No. 5, Fuxing St., Guishan Dist., Taoyuan City, 333, Taiwan

      Yu-Cheng Pei

    11. Healthy Aging Research Center, Chang Gung University, No. 259, Wenhua 1st Rd., Guishan Dist., Taoyuan City, 33302, Taiwan

      Yu-Cheng Pei

    Authors
    1. Chan-Shien Ho
      View author publications

      You can also search for this author in PubMed Google Scholar

    2. Yueh-Peng Chen
      View author publications

      You can also search for this author in PubMed Google Scholar

    3. Tzuo-Yau Fan
      View author publications

      You can also search for this author in PubMed Google Scholar

    4. Chang-Fu Kuo
      View author publications

      You can also search for this author in PubMed Google Scholar

    5. Tzu-Yun Yen
      View author publications

      You can also search for this author in PubMed Google Scholar

    6. Yuan-Chang Liu
      View author publications

      You can also search for this author in PubMed Google Scholar

    7. Yu-Cheng Pei
      View author publications

      You can also search for this author in PubMed Google Scholar

    Contributions

    C.-S.H., Y.-P.C., T.-Y.F., C.-F.K., and Y.-C.P. designed research; C.-S.H., Y.-P.C., T.-Y.F., and T.-T.Y. collected data; C.-S.H., Y.-P.C., T.-Y.F., C.-F.K., and Y.-C.P. analyzed data; C.-S.H., Y.-P.C., Y.-C.L, and Y.-C.P. wrote the paper.

    Corresponding authors

    Correspondence to Yueh-Peng Chen or Yu-Cheng Pei.

    Ethics declarations

    Conflicts of interest

    None.

    Additional information

    Publisher’s note

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

    Supplementary Information

    Below is the link to the electronic supplementary material.

    Supplementary file1 (DOCX 5379 kb)

    Rights and permissions

    Reprints and permissions

    About this article

    Check for updates. Verify currency and authenticity via CrossMark

    Cite this article

    Ho, CS., Chen, YP., Fan, TY. et al. Application of deep learning neural network in predicting bone mineral density from plain X-ray radiography. Arch Osteoporos 16, 153 (2021). https://doi.org/10.1007/s11657-021-00985-8

    Download citation

    • Received:

    • Accepted:

    • Published:

    • DOI: https://doi.org/10.1007/s11657-021-00985-8

    Keywords

    Search

    Navigation