Abstract
Summary
While FRAX with BMD could be more precise in estimating the fracture risk, DL-based models were validated to slightly reduce the number of under- and over-treated patients when no BMD measurements were available. The validated models could be used to screen for patients at a high risk of fracture and osteoporosis.
Purpose
Fracture risk assessment tool (FRAX) is useful in classifying the fracture risk level, and precise prediction can be achieved by estimating both clinical risk factors and bone mineral density (BMD) using dual X-ray absorptiometry (DXA). However, DXA is not frequently feasible because of its cost and accessibility. This study aimed to establish the reliability of deep learning (DL)-based alternative tools for screening patients at a high risk of fracture and osteoporosis.
Methods
Participants were enrolled from the National Bone Health Screening Project of Taiwan in this cross-sectional study. First, DL-based models were built to predict the lowest T-score value in either the lumbar spine, total hip, or femoral neck and their respective BMD values. The Bland–Altman analysis was used to compare the agreement between the models and DXA. Second, the predictive model to classify patients with a high fracture risk was built according to the estimated BMD from the first step and the FRAX score without BMD. The performance of the model was compared with the classification based on FRAX with BMD.
Results
Approximately 10,827 women (mean age, 65.4 ± 9.4 years) were enrolled. In the prediction of the lumbar spine BMD, total hip BMD, femoral neck BMD, and lowest T-score, the root-mean-square error (RMSE) was 0.099, 0.089, 0.076, and 0.68, respectively. The Bland–Altman analysis revealed a nonsignificant difference between the predictive models and DXA. The FRAX score with femoral neck BMD for major osteoporotic fracture risk was 9.7% ± 6.7%, whereas the risk for hip fracture was 3.3% ± 4.6%. Comparison between the classification of FRAX with and without BMD revealed the accuracy rate, positive predictive value (PPV), and negative predictive value (NPV) of 78.8%, 64.6%, and 89.9%, respectively. The area under the receiver operating characteristic curve (AUROC), accuracy rate, PPV, and NPV of the classification model were 0.913 (95% confidence interval: 0.904–0.922), 83.5%, 71.2%, and 92.2%, respectively.
Conclusion
While FRAX with BMD could be more precise in estimating the fracture risk, DL-based models were validated to slightly reduce the number of under- and over-treated patients when no BMD measurements were available. The validated models could be used to screen for patients at a high risk of fracture and osteoporosis.
Similar content being viewed by others
Data Availability
The data that support the findings of this study are available from the corresponding author, CHW, upon reasonable request.
References
Ammann P, Rizzoli R (2003) Bone strength and its determinants. Osteoporos Int 14:13–18
Sànchez-Riera L, Wilson N, Kamalaraj N, Nolla JM, Kok C, Li Y, Macara M, Norman R, Chen JS, Smith EU (2010) Osteoporosis and fragility fractures. Best Pract Res Clin Rheumatol 24:793–810
Cooper C (1997) The crippling consequences of fractures and their impact on quality of life. Am J Med 103:S12–S19
Chen I-J, Chiang C-Y, Li Y-H, Chang C-H, Hu C-C, Chen D, Chang Y, Yang W-E, Shih H-N, Ueng S-N (2015) Nationwide cohort study of hip fractures: time trends in the incidence rates and projections up to 2035. Osteoporos Int 26:681–688
Cheung EY, Tan KC, Cheung C-L, Kung AW (2016) Osteoporosis in East Asia: current issues in assessment and management. Osteoporos Sarcopenia 2:118–133
Lorentzon M (2019) Treating osteoporosis to prevent fractures: current concepts and future developments. J Intern Med 285:381–394
Dimai HP (2017) Use of dual-energy X-ray absorptiometry (DXA) for diagnosis and fracture risk assessment; WHO-criteria, T-and Z-score, and reference databases. Bone 104:39–43
Marín F, López-Bastida J, Díez-Pérez A, Sacristán J (2004) Bone mineral density referral for dual-energy X-ray absorptiometry using quantitative ultrasound as a prescreening tool in postmenopausal women from the general population: a cost-effectiveness analysis. Calcif Tissue Int 74:277–283
WHO Study Group (1994) Assessment of fracture risk and its application to screening for postmenopausal osteoporosis: report of a WHO study group. Osteoporos Int 4:368–81
Panichkul S, Sripramote M, Sriussawaamorn N (2004) Diagnostic performance of quantitative ultrasound calcaneus measurement in case finding for osteoporosis in Thai postmenopausal women. J Obstet Gynaecol Res 30:418–426
Koh LK, Sedrine WB, Torralba TP et al (2001) A simple tool to identify asian women at increased risk of osteoporosis. Osteoporos Int 12:699–705
Kanis JA, Oden A, Johansson H, Borgström F, Ström O, McCloskey E (2009) FRAX® and its applications to clinical practice. Bone 44:734–743
Hwang JS, Chan DC, Chen JF, Cheng TT, Wu CH, Soong YK, Tsai KS, Yang RS (2014) Clinical practice guidelines for the prevention and treatment of osteoporosis in Taiwan: summary. J Bone Miner Metab 32:10–16
Liu IT, Liang FW, Li CC, Chang YF, Sun ZJ, Lu TH, Chang CS, Wu CH (2022) Validation of the Taiwan FRAX® calculator for the prediction of fracture risk. Arch Osteoporos 17:27
Teeratakulpisarn N, Charoensri S, Theerakulpisut D, Pongchaiyakul C (2021) FRAX score with and without bone mineral density: a comparison and factors affecting the discordance in osteoporosis treatment in Thais. Arch Osteoporos 16:44
Alpaydin E (2021) Machine learning. MIT Press, Cambridge
Kelleher JD (2019) Deep learning. MIT press, Cambridge
Bhandary A, Prabhu GA, Rajinikanth V, Thanaraj KP, Satapathy SC, Robbins DE, Shasky C, Zhang Y-D, Tavares JMR, Raja NSM (2020) Deep-learning framework to detect lung abnormality–a study with chest X-Ray and lung CT scan images. Pattern Recogn Lett 129:271–278
Kose U, Deperlioglu O, Alzubi J, Patrut B (2021) Deep learning for medical decision support systems. Springer, Berlin
Ali F, El-Sappagh S, Islam SR, Kwak D, Ali A, Imran M, Kwak K-S (2020) A smart healthcare monitoring system for heart disease prediction based on ensemble deep learning and feature fusion. Inf Fusion 63:208–222
Chiu JS, Li YC, Yu FC, Wang YF (2006) Applying an artificial neural network to predict osteoporosis in the elderly. Stud Health Technol Inform 124:609–614
Smets J, Shevroja E, Hugle T, Leslie WD, Hans D (2021) Machine learning solutions for osteoporosis-a review. J Bone Min Res 36:833–851
Marshall D, Johnell O, Wedel H (1996) Meta-analysis of how well measures of bone mineral density predict occurrence of osteoporotic fractures. BMJ (Clinical Research ed) 312:1254–1259
Kanis JA, McCloskey EV, Johansson H, Oden A, Melton LJ III, Khaltaev N (2008) A reference standard for the description of osteoporosis. Bone 42:467–475
van der Heijden GJ, Donders AR, Stijnen T, Moons KG (2006) Imputation of missing values is superior to complete case analysis and the missing-indicator method in multivariable diagnostic research: a clinical example. J Clin Epidemiol 59:1102–1109
Azur MJ, Stuart EA, Frangakis C, Leaf PJ (2011) Multiple imputation by chained equations: what is it and how does it work? Int J Methods Psychiatr Res 20:40–49
Efron B, Tibshirani R (1997) Improvements on cross-validation: the.632+ bootstrap method. J Am Stat Assoc 92:548–560
Wu Y-C, Feng J-W (2018) Development and application of artificial neural network. Wireless Pers Commun 102:1645–1656
Baldassi C, Malatesta EM, Zecchina R (2019) Properties of the geometry of solutions and capacity of multilayer neural networks with rectified linear unit activations. Phys Rev Lett 123:170602
Arora S, Gupta A, Jain R, Nayyar A (2021) Optimization of the CNN model for hand sign language recognition using Adam optimization technique. Micro-electronics and Telecommunication Engineering. Springer, Singapore, pp 89–104
Gluer CC, Blake G, Lu Y, Blunt BA, Jergas M, Genant HK (1995) Accurate assessment of precision errors: how to measure the reproducibility of bone densitometry techniques. Osteoporos Int 5:262–270
Chen R-C, Caraka RE, Arnita NEG et al (2020) An end to end of scalable tree boosting system. Sylwan 165:1–11
Bland JM, Altman DG (1986) Statistical methods for assessing agreement between two methods of clinical measurement. Lancet (London, England) 1:307–310
Fluss R, Faraggi D, Reiser B (2005) Estimation of the Youden index and its associated cutoff point. Biom J Biom Z 47:458–472
Dagan N, Elnekave E (2020) Automated opportunistic osteoporotic fracture risk assessment using computed tomography scans to aid in FRAX underutilization. Nat Med 26:77–82
Cadarette SM, Jaglal SB, Kreiger N, McIsaac WJ, Darlington GA, Tu JV (2000) Development and validation of the osteoporosis risk assessment instrument to facilitate selection of women for bone densitometry. CMAJ Can Med Assoc J 162:1289–1294
Lydick E, Cook K, Turpin J, Melton M, Stine R, Byrnes C (1998) Development and validation of a simple questionnaire to facilitate identification of women likely to have low bone density. Am J Manag Care 4:37–48
Sedrine WB, Chevallier T, Zegels B, Kvasz A, Micheletti MC, Gelas B, Reginster JY (2002) Development and assessment of the osteoporosis index of risk (OSIRIS) to facilitate selection of women for bone densitometry. Gynecol Endocrinol 16:245–250
Dane C, Dane B, Cetin A, Erginbas M (2008) The role of quantitative ultrasound in predicting osteoporosis defined by dual-energy X-ray absorptiometry in pre- and postmenopausal women. Climacteric 11:296–303
Hsieh CI, Zheng K, Lin C et al (2021) Automated bone mineral density prediction and fracture risk assessment using plain radiographs via deep learning. Nat Commun 12:5472
Wu Q, Nasoz F, Jung J, Bhattarai B, Han MV, Greenes RA, Saag KG (2021) Machine learning approaches for the prediction of bone mineral density by using genomic and phenotypic data of 5130 older men. Sci Rep 11:4482
Shioji M, Yamamoto T, Ibata T, Tsuda T, Adachi K, Yoshimura N (2017) Artificial neural networks to predict future bone mineral density and bone loss rate in Japanese postmenopausal women. BMC Res Notes 10:590
Fan B, Lu Y, Genant H, Fuerst T, Shepherd J (2010) Does standardized BMD still remove differences between Hologic and GE-Lunar state-of-the-art DXA systems? Osteoporos Int 21:1227–1236
Acknowledgements
The Taiwan Osteoporosis Association (TOA) conducted the national circuit program for BMD measurements, sponsored by Merck Sharp & Dohme Pharmaceutical Company. A bus installed with a DXA machine and a nurse and a radiology technician, both certified by the International Society of Clinical Densitometry, were sponsored by Merck Sharp & Dohme Pharmaceutical Company. E-Da hospital, the TOA, and Merck Sharp & Dohme Pharmaceutical Company had no role in the design and conduct of the study; collection, management, analysis, and interpretation of the data; preparation, review, or approval of the manuscript; or the decision to submit the manuscript for publication.
Funding
This study was supported by the research project of E-Da Hospital, Taiwan (grant number: EDAHP110007).
Author information
Authors and Affiliations
Corresponding authors
Ethics declarations
Ethical approval
This study was approved by the local institutional review board of Chang Gung Memorial Hospital (102-1878B) and the institutional review board of E-Da Hospital (EMRP-110–085). Informed consent was obtained from all individuals who participated in the study. 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.
Conflicts of interest
None.
Additional information
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Rights and permissions
Springer Nature or its licensor (e.g. a society or other partner) holds exclusive rights to this article under a publishing agreement with the author(s) or other rightsholder(s); author self-archiving of the accepted manuscript version of this article is solely governed by the terms of such publishing agreement and applicable law.
About this article
Cite this article
Hung, W.C., Lin, YL., Cheng, TT. et al. Establish and validate the reliability of predictive models in bone mineral density by deep learning as examination tool for women. Osteoporos Int 35, 129–141 (2024). https://doi.org/10.1007/s00198-023-06913-5
Received:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s00198-023-06913-5