Risk prediction of type 2 diabetes in steel workers based on convolutional neural network

  • Jian-Hui Wu
  • Jing Li
  • Jie Wang
  • Lu Zhang
  • Hai-Dong Wang
  • Guo-Li Wang
  • Xiao-lin Li
  • Ju-Xiang YuanEmail author
Original Article


With the change in environment and lifestyle, the number of diabetic patients is increasing rapidly. Diabetes has become one of the most important chronic diseases affecting the health of the Chinese people, and complications, disability, death and treatment costs caused by diabetes have placed a heavy burden on families and society. If the high-risk population of diabetes can be identified and the adverse lifestyle can be changed as soon as possible, the incidence of diabetes can be reduced or the onset of diabetes can be slowed down. Risk prediction model can accurately predict the risk of disease and has been widely used in the field of health management and medical care. This study was based on the special occupational group of steel workers, the risk prediction model of type 2 diabetes was established by using convolutional neural network, and the feasibility of the model was discussed. The results showed that the prediction accuracy of the established model in the training set, verification set, and test set is relatively high, which is 94.5%, 91.0%, and 89.0%, respectively. The area under the ROC curve was 0.950 (95 CI 0.938–0.962), 0.916 (95 CI 0.888–0.945), and 0.899 (95 CI 0.899–0.939), respectively, indicating that the model can accurately predict the risk of type 2 diabetes among steel workers, provide a basis for self-health management of steel workers, facilitate the rational allocation of medical and health resources and the development of health services, and provide a basis for government departments to make decisions.


Convolutional neural network Steel worker Type 2 diabetes mellitus (T2DM) Prediction of morbidity risk 



This work was supported by the National Key R&D Program of China (No. 2016YFC0900605).

Compliance with ethical standards

Conflict of interest

The authors declare that there are no conflicts of interest regarding the publication of this paper.


  1. 1.
    Colman PG, Mcnair PD, Gellert S (2010) Development of autoantibodies to islet antigens during childhood: implications for preclinical type 1 diabetes screening. Pediatric Diabetes 3(3):144–148Google Scholar
  2. 2.
    Mi S-Q (2011) Establishing and validating of type II diabetes incident prediction model of Chinese adult at individual level. Ph.D. dissertation, Chronic center, Chinese Center for Disease Control and Prevention, Beijing, ChinaGoogle Scholar
  3. 3.
    Hong Y (2016) Research on diabetes prediction models based on machine learning algorithm. M.S. thesis, School of Management, Harbin Institute of Technology, Harbin, ChinaGoogle Scholar
  4. 4.
    Nichols GA, Brown JB (2008) Validating the framingham offspring study equations for predicting incident diabetes mellitus. Am J Managed Care 14(9):574–580Google Scholar
  5. 5.
    Stern Michael P (2002) Identification of persons at high risk for type 2 diabetes mellitus: Do we need the oral glucose tolerance test? Ann Intern Med 136(8):575–581Google Scholar
  6. 6.
    McNeely MJ, Boyko EJ, Leonetti DL, Kahn SE, Fujimoto WY (2003) Comparison of a clinical model, the oral glucose tolerance test, and fasting glucose for prediction of type 2 diabetes risk in Japanese Americans. Diabetes Care 26(3):758–763Google Scholar
  7. 7.
    Han Y, Li J, Yang XL, Liu WX, Zhang YZ (2018) Dynamic prediction research of silicon content in hot metal driven by big data in blast furnace smelting process under hadoop cloud platform. Complexity 2018:1–16Google Scholar
  8. 8.
    Liu FC, Liu YL, Jin DH, Jia XY, Wang TT (2018) Research on workshop-based positioning technology based on internet of things in big data background. Complexity 2018:1–11Google Scholar
  9. 9.
    Yang AM, Yang XL, Chang JC, Bai B, Kong FB, Ran QB (2018) Research on a fusion scheme of cellular network and wireless sensor networks for cyber physical social systems. IEEE Access 6(99):18786–18794Google Scholar
  10. 10.
    Hai-Yun Wu, Pan Ping, He Yao, Liu Xiao-Ling, Zeng Qiang, Wei Cheng-Jun, Huang Shi-Jian (2007) A risk model for prediction of diabetes in Chinese adults. Chin J Health Manag 1(2):95–98Google Scholar
  11. 11.
    Chien K, Cai T, Hsu H, Su T, Chang W, Chen M (2009) A prediction model for type 2 diabetes risk among Chinese people. Diabetologia 52(3):443–450Google Scholar
  12. 12.
    Siyouneh B, Hsuan-Ping L, Richard P, Melanie M, Martha S, Bradlee M (2018) Dietary cholesterol intake is not associated with risk of type 2 diabetes in the Framingham offspring study. Nutrients 10(6):1–9Google Scholar
  13. 13.
    Hippisley-Cox J, Coupland C (2010) Individualizing the risks of statins in men and women in England and Wales: population-based cohort study. Heart 96(12):939–947Google Scholar
  14. 14.
    Wu JH, Wei W, Zhang L (2019) Risk assessment of hypertension in steel workers based on LVQ and fisher-SVM deep excavation. IEEE Access 7(1):23109–23119Google Scholar
  15. 15.
    Zhang Y, Zhao CK, Li XL, He YP, Ren WW, Zou CP (2017) Virtual touch tissue imaging and quantification: value in malignancy prediction for complex cystic and solid breast lesions. Sci Rep 7(1):1–10Google Scholar
  16. 16.
    Oyedotun OK, Khashman A (2016) Document segmentation using textural features summarization and feedforward neural network. Appl Intell 45(1):1–15Google Scholar
  17. 17.
    Konomi M, Sacha GM (2017) Feedforward neural network methodology to characterize thin films by electrostatic force microscopy. Ultramicroscopy 182:243–248Google Scholar
  18. 18.
    Kalchbrenner N, Grefenstette E, Blunsom P (2014) A convolutional neural network for modelling sentences. Eprint Arxiv, 655–665Google Scholar
  19. 19.
    Jeon J, Paik J (2015) Single image super-resolution based on subpixel shifting model. Optik Int J Light Electron Optics 126(24):4954–4959Google Scholar
  20. 20.
    Herrera C, Nava FA (2009) Earthquake hazard assessment in seismogenic systems through Markovian artificial neural network estimation: an application to the Japan area. Earth Planets Space 61(11):1223–1232Google Scholar
  21. 21.
    Xu R, Chen T, Xia Y, Lu Q, Liu B, Wang X (2015) Word embedding composition for data imbalances in sentiment and emotion classification. Cognit Comput 7(2):226–240Google Scholar
  22. 22.
    He F, Zhang L (2018) Prediction model of end-point phosphorus content in BOF steelmaking process based on PCA and BP neural network. J Process Control 66:51–58Google Scholar
  23. 23.
    Feng Yu, Xiaozhong Xu (2014) A short-term load forecasting model of natural gas based on optimized genetic algorithm and improved BP neural network. Appl Energy 134:102–113Google Scholar
  24. 24.
    Adler J, Oktem O (2018) Learned Primal-dual Reconstruction. IEEE Trans Med Imaging 2018:1–10Google Scholar
  25. 25.
    Agostinelli F, Hoffman M, Sadowski P, Baldi P (2014) Learning activation functions to improve deep neural networks. Comput Sci 1–9. ArXiv preprint arXiv:1412.6830
  26. 26.
    Wang Y, Wang X, Song G, Liu H, Zhu D, Ding J (2016) Genetic connection between mud shale lithofacies and shale oil enrichment in Jiyang depression, Bohai bay basin. Pet Explor Dev 43(5):759–768Google Scholar
  27. 27.
    Yang AM, Li SS, Lin HL, Jin DH (2018) Edge extraction of mineralogical phase based on fractal theory. Chaos Solitions Fractals 117:215–221Google Scholar
  28. 28.
    Kim J, Kim J, Jang GJ, Lee M (2017) Fast learning method for convolutional neural networks using extreme learning machine and its application to lane detection. Neural Netw 87:109–121Google Scholar
  29. 29.
    World Health Organization (2006) Definition and diagnosis of diabetes mellitus and intermediate hyperglycemia: report of a WHO/IDF consultation. Geneva World Health OrganizationGoogle Scholar

Copyright information

© Springer-Verlag London Ltd., part of Springer Nature 2019

Authors and Affiliations

  1. 1.School of Public HealthNorth China University of Science and TechnologyTangshanPeople’s Republic of China
  2. 2.Hebei Province Key Laboratory of Occupational Health and Safety for Coal IndustryNorth China University of Science and TechnologyTangshanPeople’s Republic of China
  3. 3.INSA Centre Val de LoireBlois CedexFrance

Personalised recommendations