Abstract
Osteoporosis is a common disease that increases fracture risk. Hip fractures, especially in elderly people, lead to increased morbidity, decreased quality of life and increased mortality. Being a silent disease before fracture, osteoporosis often remains undiagnosed and untreated. Areal bone mineral density (aBMD) assessed by dual-energy X-ray absorptiometry (DXA) is the gold-standard method for osteoporosis diagnosis and hence also for future fracture prediction (prognostic). However, the required special equipment is not broadly available everywhere, in particular not to patients in developing countries. We propose a deep learning classification model (FORM) that can directly predict hip fracture risk from either plain radiographs (X-ray) or 2D projection images of computed tomography (CT) data. Our method is fully automated and therefore well suited for opportunistic screening settings, identifying high risk patients in a broader population without additional screening. FORM was trained and evaluated on X-rays and CT projections from the Osteoporosis in Men (MrOS) study. 3108 X-rays (89 incident hip fractures) or 2150 CTs (80 incident hip fractures) with a 80/20 split (training/validation) were used. We show that FORM can correctly predict the 10-year hip fracture risk with a validation AUC of 81.44% ± 3.11%/81.04% ± 5.54% (mean ± STD) including additional information like age, BMI, fall history and health background across a 5-fold cross validation on the X-ray and CT cohort, respectively. Our approach significantly (p < 0.01) outperforms previous methods like Cox Proportional-Hazards Model and
with 70.19 ± 6.58 and 74.72 ± 7.21 respectively on the X-ray cohort. Our model outperform on both cohorts hip aBMD based predictions (validation AUC 82.67% ± 0.21% vs. 71.82% ± 0.50% and 78.41% ± 0.33 vs. 76.55% ± 0.89%). We are confident that FORM can contribute on improving osteoporosis diagnosis at an early stage.
This is a preview of subscription content, log in via an institution to check access.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Similar content being viewed by others
Notes
- 1.
The Osteoporotic Fractures in Men (MrOS) Study: https://mrosonline.ucsf.edu.
- 2.
The Study of Osteoporotic Fractures (SOF:) https://sofonline.ucsf.edu.
- 3.
Example image and key points only for illustrative purpose; image source https://radiopaedia.org/cases/normal-hip-x-rays.
- 4.
In the MrOS study, the phantoms are used to calibrate HU to BMD. In this work no BMD calibration is performed for a more realistic opportunistic screening setting.
- 5.
Implemented in Tensorflow 2.4, source code will be release on publication, experiments executed on Nvidia RTX 3090, inference < 1 s per image.
- 6.
https://mrosonline.ucsf.edu, Update august 2021.
References
Bishop, C.M.: Neural Networks for Pattern Recognition. Oxford University Press, Inc., Oxford (1996)
Black, D.M., et al.: Proximal femoral structure and the prediction of hip fracture in men: a large prospective study using QCT. J. Bone Mineral Res. 23(8), 1326–1333 (2008)
Bredbenner, T.L., Mason, R.L., Havill, L.M., Orwoll, E.S., Nicolella, D.P., Osteoporotic Fractures in Men (MrOS) Study: Fracture risk predictions based on statistical shape and density modeling of the proximal femur. J. Bone Mineral Res. 29(9), 2090–2100 (2014)
Camacho, P.M., et al.: American Association of Clinical Endocrinologists/American College of Endocrinology clinical practice guidelines for the diagnosis and treatment of postmenopausal osteoporosis-2020 update. Endocr. Pract. 26, 1–46 (2020)
Cox, D.R.: Regression models and life-tables. J. R. Stat. Soc. Ser. B (Methodol.) 34(2), 187–202 (1972)
Damm, T., et al.: Artificial intelligence-driven hip fracture prediction based on pelvic radiographs exceeds performance of DXA: the “Study of Osteoporotic Fractures’’ (SOF). J. Bone Mineral Res. 37, 193 (2021)
Ebeling, P.R.: Osteoporosis in men: why change needs to happen. World Osteoporosis Day Thematic Report International Osteoporosis Foundation, Nyon (2014)
Pearson, K.: LIII. On lines and planes of closest fit to systems of points in space. Lond. Edinb. Dublin Philos. Mag. J. Sci. 2(11), 559–572 (1901)
Goodfellow, I., Bengio, Y., Courville, A.: Deep Learning. MIT Press, Cambridge (2016)
Grossmann, V., Schmarje, L., Koch, R.: Beyond hard labels: investigating data label distributions. arXiv preprint arXiv:2207.06224 (2022)
Hamidi, Z.: What’s BMD and What We Do in a BMD Centre?, pp. 225–246 (2012)
He, K., Zhang, X., Ren, S., Sun, J.: Deep residual learning for image recognition. In: IEEE Conference on Computer Vision and Pattern Recognition (CVPR), pp. 770–778 (2015)
Hippisley-Cox, J., Coupland, C.: Predicting risk of osteoporotic fracture in men and women in England and Wales: prospective derivation and validation of QFractureScores. BMJ 339, b4229 (2009)
Ho, C.-S., et al.: Application of deep learning neural network in predicting bone mineral density from plain X-ray radiography. Arch. Osteoporos. 16(1), 1–12 (2021). https://doi.org/10.1007/s11657-021-00985-8
Hsieh, C.I., et al.: Automated bone mineral density prediction and fracture risk assessment using plain radiographs via deep learning. Nat. Commun. 12(1), 1–9 (2021)
Jang, R., Choi, J.H., Kim, N., Chang, J.S., Yoon, P.W., Kim, C.H.: Prediction of osteoporosis from simple hip radiography using deep learning algorithm. Sci. Rep. 11(1), 1–9 (2021)
Johnell, O., Kanis, J.A.: An estimate of the worldwide prevalence and disability associated with osteoporotic fractures. Osteoporos. Int. 17(12), 1726–1733 (2006)
Kanis, J.A., Johnell, O., Odén, A., Johansson, H., McCloskey, E.: FRAX™ and the assessment of fracture probability in men and women from the UK. Osteoporos. Int. 19(4), 385–397 (2008)
Krizhevsky, A., Sutskever, I., Hinton, G.E.: ImageNet classification with deep convolutional neural networks. In: Advances in Neural Information Processing Systems, vol. 60, pp. 1097–1105. Association for Computing Machinery (2012)
Langsetmo, L., et al.: Volumetric bone mineral density and failure load of distal limbs predict incident clinical fracture independent of FRAX and clinical risk factors among older men. J. Bone Mineral Res. 33(7), 1302–1311 (2018)
Pickhardt, P.J., Pooler, B.D., Lauder, T., del Rio, A.M., Bruce, R.J., Binkley, N.: Opportunistic screening for osteoporosis using abdominal computed tomography scans obtained for other indications. Ann. Intern. Med. 158(8), 588–595 (2013)
Prasad, D., Nguyen, M.H.: Chronic hepatitis, osteoporosis, and men: under-recognised and underdiagnosed. Lancet Diabetes Endocrinol. 9(3), 141 (2021)
Salari, N., et al.: The global prevalence of osteoporosis in the world: a comprehensive systematic review and meta-analysis. J. Orthop. Surg. Res. 16(1), 1–20 (2021)
Santarossa, M., et al.: MedRegNet: unsupervised multimodal retinal-image registration with GANs and ranking loss. In: Medical Imaging 2022: Image Processing, vol. 12032, pp. 321–333. SPIE (2022)
Schmarje, L., Brünger, J., Santarossa, M., Schröder, S.M., Kiko, R., Koch, R.: Fuzzy overclustering: semi-supervised classification of fuzzy labels with overclustering and inverse cross-entropy. Sensors 21(19), 6661 (2021)
Schmarje, L., et al.: Is one annotation enough? A data-centric image classification benchmark for noisy and ambiguous label estimation. arXiv preprint arXiv:2207.06214 (2022)
Schmarje, L., et al.: A data-centric approach for improving ambiguous labels with combined semi-supervised classification and clustering. arXiv preprint arXiv:2106.16209 (2022)
Schousboe, J.T., et al.: Prediction models of prevalent radiographic vertebral fractures among older men. J. Clin. Densitom. 17(4), 449–457 (2014)
Schousboe, J.T., et al.: Prediction of incident major osteoporotic and hip fractures by trabecular bone score (TBS) and prevalent radiographic vertebral fracture in older men. J. Bone Mineral Res. 31(3), 690–697 (2016)
Schousboe, J.T., et al.: Predictors of change of trabecular bone score (TBS) in older men: results from the Osteoporotic Fractures in Men (MrOS) Study. Osteoporos. Int. 29(1), 49–59 (2017). https://doi.org/10.1007/s00198-017-4273-z
Sohn, K., et al.: FixMatch: simplifying semi-supervised learning with consistency and confidence. In: Advances in Neural Information Processing Systems 33 Pre-proceedings (NeurIPS 2020) (2020)
Su, Y., Kwok, T.C.Y., Cummings, S.R., Yip, B.H.K., Cawthon, P.M.: Can classification and regression tree analysis help identify clinically meaningful risk groups for hip fracture prediction in older American men (the MrOS cohort study)? JBMR Plus 3(10), e10207 (2019)
Treece, G.M., Osteoporotic Fractures in Men (MrOS) Study, et al.: Predicting hip fracture type with cortical bone mapping (CBM) in the osteoporotic fractures in men (MrOS) study. J. Bone Mineral Res. 30(11), 2067–2077 (2015)
US Preventive Services Task Force: Screening for osteoporosis: US preventive services task force recommendation statement. Ann. Intern. Med. 154(5), 356–364 (2011)
Wang, X., et al.: Prediction of new clinical vertebral fractures in elderly men using finite element analysis of CT scans. J. Bone Mineral Res. 27(4), 808–816 (2012)
Watts, N.B.: The fracture risk assessment tool (FRAX®): applications in clinical practice. J. Women’s Health 20(4), 525–531 (2011)
Welch, B.L.: The generalization of ‘student’s’ problem with several different population variances are involved. Biometrika 34(1–2), 28–35 (1947)
Yamamoto, N., et al.: Deep learning for osteoporosis classification using hip radiographs and patient clinical covariates. Biomolecules 10(11), 1534 (2020)
Yang, L., Parimi, N., Orwoll, E.S., Black, D.M., Schousboe, J.T., Eastell, R.: Association of incident hip fracture with the estimated femoral strength by finite element analysis of DXA scans in the Osteoporotic Fractures in Men (MrOS) study. Osteoporos. Int. 29(3), 643–651 (2017). https://doi.org/10.1007/s00198-017-4319-2
Acknowledgements
We acknowledge funding of Lars Schmarje, Stefan Reinhold, Timo Damm and Claus C. Glüer by the ARTEMIS project (grant no. 01EC1908E) funded by the Federal Ministry of Education and Research (BMBF), Germany.
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
1 Electronic supplementary material
Below is the link to the electronic supplementary material.
Rights and permissions
Copyright information
© 2022 The Author(s), under exclusive license to Springer Nature Switzerland AG
About this paper
Cite this paper
Schmarje, L., Reinhold, S., Damm, T., Orwoll, E., Glüer, CC., Koch, R. (2022). Opportunistic Hip Fracture Risk Prediction in Men from X-ray: Findings from the Osteoporosis in Men (MrOS) Study. In: Rekik, I., Adeli, E., Park, S.H., Cintas, C. (eds) Predictive Intelligence in Medicine. PRIME 2022. Lecture Notes in Computer Science, vol 13564. Springer, Cham. https://doi.org/10.1007/978-3-031-16919-9_10
Download citation
DOI: https://doi.org/10.1007/978-3-031-16919-9_10
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-031-16918-2
Online ISBN: 978-3-031-16919-9
eBook Packages: Computer ScienceComputer Science (R0)
