Comparing Machine Learning Algorithms for Prediction of Osteoporosis
A Systematic Review and Meta-Analysis Study
Keywords:
osteoporosis, prediction , machine learning, algorithmAbstract
Introduction: Osteoporosis is a prevalent bone disease that affects millions of individuals worldwide. Early identification and prediction of osteoporosis can enable timely interventions and preventive measures. This study investigates the potential of machine learning algorithms to accurately predict osteoporosis.
Material and Methods: This study is a systematic review and meta-analysis conducted by searching in three databases. Our search encompassed databases such as Web of Science, PubMed, and Scopus. Pertinent information from the selected studies was independently extracted by two authors. The PRISMA guidelines were followed to ensure a rigorous review process. The PROBAST tool was utilized to assess the risk of bias in the included studies. Data analysis was performed using Stata (v.17.1).
Results: A total of 63 algorithms from 18 studies were evaluated. In terms of predicting osteoporosis, support vector machine (SVM) and random forest (RF) algorithms demonstrated the highest sensitivity. For SVM, the sensitivity and diagnostic odds ratio (DOR) were 83.0% (95% confidence interval (CI): 76.0-88.0) and 10.4 (95% CI: 6.0-18.2), respectively. Similarly, in the case of RF algorithm, the sensitivity and DOR were 81.0% (95% CI: 74.0-87.0) and 13.0 (95% CI: 7.7-21.2), respectively. The artificial neoural networks (ANN), RF, and K-nearest neighbors (KNN) algorithms exhibited the highest specificity values: ANN- specificity of 79.0% (95% CI: 71.0-85.0) and DOR of 12.0 (7.3-18.7); RF- specificity of 75.0% (95% CI: 62.0-84.0) and DOR of 13.0 (7.7-21.2); KNN- specificity of 75.0% (95% CI: 67.0-82.0) and DOR of 7.7(6.6-9.0).
Conclusion: Our study highlights the promising potential of machine learning algorithms for the accurate prediction of osteoporosis. ANN model and SVM, RF, and KNN algorithms have emerged as the most robust predictors. These findings demonstrate substantial potential for aiding early detection and intervention strategies against osteoporosis.
References
1. Mirza F, Canalis E. Management of endocrine disease: Secondary osteoporosis: Pathophysiology and management. Eur J Endocrinol. 2015; 173(3): R131-51. PMID: 25971649 DOI: 10.1530/EJE-15-0118
2. Gulsahi A. Osteoporosis and jawbones in women. J Int Soc Prev Community Dent. 2015; 5(4): 263-7. PMID: 26312225 DOI: 10.4103/2231-0762.161753
3. Cole ZA, Dennison EM, Cooper C. Osteoporosis epidemiology update. Curr Rheumatol Rep. 2008; 10(2): 92-6. PMID: 18460262 DOI: 10.1007/s11926-008-0017-6
4. Sinaki M. Critical appraisal of physical rehabilitation measures after osteoporotic vertebral fracture. Osteoporos Int. 2003; 14(9): 773-9. PMID: 12904834 DOI: 10.1007/s00198-003-1446-8
5. Miyakoshi N, Itoi E, Kobayashi M, Kodama H. Impact of postural deformities and spinal mobility on quality of life in postmenopausal osteoporosis. Osteoporos Int. 2003; 14(12): 1007-12. PMID: 14557854 DOI: 10.1007/s00198-003-1510-4
6. Sozen T, Ozısık L, Basaran NC. An overview and management of osteoporosis. Eur J Rheumatol. 2017; 4(1): 46-56. PMID: 28293453 DOI: 10.5152/eurjrheum.2016.048
7. Lippuner K, Rimmer G, Stuck AK, Schwab P, Bock O. Hospitalizations for major osteoporotic fractures in Switzerland: A long-term trend analysis between 1998 and 2018. Osteoporos Int. 2022; 33(11): 2327-35. PMID: 35916908 DOI: 10.1007/s00198-022-06481-0
8. Sukegawa S, Fujimura A, Taguchi A, Yamamoto N, Kitamura A, Goto R, et al. Identification of osteoporosis using ensemble deep learning model with panoramic radiographs and clinical covariates. Sci Rep. 2022; 12(1): 6088. PMID: 35413983 DOI: 10.1038/s41598-022-10150-x
9. Ponnusamy KE, Iyer S, Gupta G, Khanna A. Instrumentation of the osteoporotic spine: Biomechanical and clinical considerations. Spine J. 2011; 11(1): 54-63. PMID: 21168099 DOI: 10.1016/j.spinee.2010.09.024
10. Dempster DW. Osteoporosis and the burden of osteoporosis-related fractures. Am J Manag Care. 2011; 17(Suppl 6): S164-9. PMID: 21761955
11. Tejaswini E, Vaishnavi P, Sunitha R. Detection and prediction of osteoporosis using impulse response technique and artificial neural network. International Conference on Advances in Computing, Communications and Informatics. IEEE; 2016.
12. Anam M, Ponnusamy V, Hussain M, Nadeem MW, Javed M, Goh HG, et al. Osteoporosis prediction for trabecular bone using machine learning: A review. Computers, Materials and Continua. 2020; 67(1): 89-105.
13. De Vries BCS, Hegeman JH, Nijmeijer W, Geerdink J, Seifert C, Groothuis-Oudshoorn CGM. Comparing three machine learning approaches to design a risk assessment tool for future fractures: Predicting a subsequent major osteoporotic fracture in fracture patients with osteopenia and osteoporosis. Osteoporos Int. 2021; 32(3): 437-49. PMID: 33415373 DOI: 10.1007/s00198-020-05735-z
14. Shim J-G, Kim DW, Ryu K-H, Cho EA, Ahn JH, Kim JI, et al. Application of machine learning approaches for osteoporosis risk prediction in postmenopausal women. Arch Osteoporos. 2020; 15(1): 169. PMID: 33097976 DOI: 10.1007/s11657-020-00802-8
15. Moons KGM, Wolff RF, Riley RD, Whiting PF, Westwood M, Collins GS, et al. PROBAST: A tool to assess the risk of bias and applicability of prediction model studies. Ann Intern Med. 2019; 170(1): W1-33. PMID: 30596876 DOI: 10.7326/M18-1377
16. Ordonez C, Matias JM, Juez JFD, Garcia PJ. Machine learning techniques applied to the determination of osteoporosis incidence in post-menopausal women. Mathematical and Computer Modelling. 2009; 50: 673-9.
17. Anastassopoulos G, Adamopoulos A, Galiatsatos D, Drosos G. Osteoporosis risk factor estimation using artificial neural networks and genetic algorithms. Stud Health Technol Inform. 2013: 190: 186-8. PMID: 23823417
18. Chang HW, Chiu YH, Kao HY, Yang CH, Ho WH. Comparison of classification algorithms with wrapper-based feature selection for predicting osteoporosis outcome based on genetic factors in a Taiwanese women population. Int J Endocrinol. 2013; 2013: 850735. PMID: 23401685 DOI: 10.1155/2013/850735
19. Kim SK, Yoo TK, Oh E, Kim DW. Osteoporosis risk prediction using machine learning and conventional methods. Annu Int Conf IEEE Eng Med Biol Soc. 2013; 2013: 188-91. PMID: 24109656 DOI: 10.1109/EMBC.2013.6609469
20. Yoo TK, Kim SK, Kim DW, Choi JY, Lee WH, Oh E, et al. Osteoporosis risk prediction for bone mineral density assessment of postmenopausal women using machine learning. Yonsei Med J. 2013; 54(6): 1321-30. PMID: 24142634 DOI: 10.3349/ymj.2013.54.6.1321
21. Yu XH, Ye C, Xiang L. Application of artificial neural network in the diagnostic system of osteoporosis. Neurocomputing. 2016; 214: 376-81.
22. Meng J, Sun N, Chen YL, Li Z, Cui X, Fan J, et al. Artificial neural network optimizes self-examination of osteoporosis risk in women. J Int Med Res. 2019; 47(7): 3088-98. PMID: 31179797 DOI: 10.1177/0300060519850648
23. Yang WYO, Lai CC, Tsou MT, Hwang LC. Development of machine learning models for prediction of osteoporosis from clinical health examination data. Int J Environ Res Public Health. 2021; 18(14): 7635. PMID: 34300086 DOI: 10.3390/ijerph18147635
24. Wang YQ, Wang LX, Sun YL, Wu M, Ma Y, Yang L, et al. Prediction model for the risk of osteoporosis incorporating factors of disease history and living habits in physical examination of population in Chongqing, Southwest China: Based on artificial neural network. BMC Public Health. 2021; 21(1): 991. PMID: 34039329 DOI: 10.1186/s12889-021-11002-5
25. Lim HK, Ha HI, Park SY, Han J. Prediction of femoral osteoporosis using machine-learning analysis with radiomics features and abdomen-pelvic CT: A retrospective single center preliminary study. PLoS One. 2021; 16(3): e0247330. PMID: 33661911 DOI: 10.1371/journal.pone.0247330
26. Klontzas ME, Manikis GC, Nikiforaki K, Vassalou EE, Spanakis K, Stathis I, et al. Radiomics and machine learning can differentiate transient osteoporosis from avascular necrosis of the hip. Diagnostics (Basel). 2021; 11(9): 1686. PMID: 34574027 DOI: 10.3390/diagnostics11091686
27. Patil KA, Prashanth KVM, Ramalingaiah A. Classification of osteoporosis in the lumbar vertebrae using L2 regularized neural network based on PHOG features. International Journal of Advanced Computer Science and Applications. 2022; 13(4): 413-23.
28. Huang CB, Hu JS, Tan K, Zhang W, Xu TH, Yang L. Application of machine learning model to predict osteoporosis based on abdominal computed tomography images of the psoas muscle: A retrospective study. BMC Geriatr. 2022; 22(1): 796. PMID: 36229793 DOI: 10.1186/s12877-022-03502-9
29. Zeitlin J, Parides MK, Lane JM, Russell LA, Kunze KN. A clinical prediction model for 10-year risk of self-reported osteoporosis diagnosis in pre- and perimenopausal women. Arch Osteoporos. 2023; 18(1): 78. PMID: 37273115 DOI: 10.1007/s11657-023-01292-0
30. Wu X, Zhai F, Chang A, Wei J, Guo Y, Zhang J. Application of machine learning algorithms to predict osteoporosis in postmenopausal women with type 2 diabetes mellitus. J Endocrinol Invest. 2023; 46(12): 2535-46. PMID: 37171784 DOI: 10.1007/s40618-023-02109-0
31. Sebro R, Elmahdy M. Machine learning for opportunistic screening for osteoporosis and osteopenia using knee CT scans. Can Assoc Radiol J. 2023; 74(4): 676-87. PMID: 36960893 DOI: 10.1177/08465371231164743
32. Lin Y-T, Chu C-Y, Hung K-S, Lu C-H, Bednarczyk EM, Chen H-Y. Can machine learning predict pharmacotherapy outcomes? An application study in osteoporosis. Comput Methods Programs Biomed. 2022; 225: 107028. PMID: 35930862 DOI: 10.1016/j.cmpb.2022.107028
33. Fasihi L, Tartibian B, Eslami R, Fasihi H. Artificial intelligence used to diagnose osteoporosis from risk factors in clinical data and proposing sports protocols. Sci Rep. 2022; 12(1): 18330. PMID: 36316387 DOI: 10.1038/s41598-022-23184-y
34. Maroco J, Silva D, Rodrigues A, Guerreiro M, Santana I, de Mendonca A. Data mining methods in the prediction of Dementia: A real-data comparison of the accuracy, sensitivity and specificity of linear discriminant analysis, logistic regression, neural networks, support vector machines, classification trees and random forests. BMC Res Notes. 2011; 4: 299. PMID: 21849043 DOI: 10.1186/1756-0500-4-299
35. Yao Y, Liu Y, Yu Y, Xu H, Lv W, Li Z, et al. K-SVM: An effective SVM algorithm based on K-means clustering. Journal of Computers. 2013; 8(10): 2632-9.
36. Qiu X, Zhang L, Suganthan PN, Amaratunga GAJ. Oblique random forest ensemble via least square estimation for time series forecasting. Information Sciences. 2017; 420: 249-62.
37. Javaid M, Haleem A, Singh RP, Suman R, Rab S. Significance of machine learning in healthcare: Features, pillars and applications. International Journal of Intelligent Networks. 2022; 3: 58-73.
38. Belgiu M, Dragut L. Random forest in remote sensing: A review of applications and future directions. ISPRS Journal of Photogrammetry and Remote Sensing. 2016; 114: 24-31.
39. Saputra M, Mawengkang H, Nababan E. Gini index with local mean based for determining k value in k-nearest neighbor classification. Journal of Physics: Conference Series. 2019; 1235: 012006.
40. Thawnashom K, Pornsawad P, Makond B. Machine learning's performance in classifying postmenopausal osteoporosis Thai patients. Intelligence-Based Medicine. 2023; 7: 100099.
Published
Issue
Section
License
Copyright (c) 2026 Advances in Medical Informatics
This work is licensed under a Creative Commons Attribution-NonCommercial 4.0 International License.
