Healthc Inform Res.  2017 Oct;23(4):285-292. 10.4258/hir.2017.23.4.285.

Development of a Stress Classification Model Using Deep Belief Networks for Stress Monitoring

• Song SH
• Kim DK

• KMID: 2396420
• DOI: http://doi.org/10.4258/hir.2017.23.4.285

Abstract

OBJECTIVES
Stress management is related to public healthcare and quality of life; an accurate stress classification method is necessary for the design of stress monitoring systems. Therefore, the goal of this study was to design a novel stress classification model using a deep learning method.
METHODS
In this paper, we present a stress classification model using the dataset from the sixth Korea National Health and Nutrition Examination Survey conducted from 2013 to 2015 (KNHANES VI) to analyze stress-related health data. Statistical analysis was performed to identify the nine features of stress detection, and we evaluated the performance of the proposed stress classification by comparison with several stress detection models. The proposed model was also evaluated using Deep Belief Networks (DBN).
RESULTS
We designed profiles depending on the number of hidden layers, nodes, and hyper-parameters according to the loss function results. The experimental results showed that the proposed model achieved an accuracy and a specificity of 66.23% and 75.32%, respectively. The proposed DBN model performed better than other classification models, such as support vector machine, naive Bayesian classifier, and random forest.
CONCLUSIONS
The proposed model in this study was demonstrated to be effective in classifying stress detection, and in particular, it is expected to be applicable for stress prediction in stress monitoring systems.

Keyword

Stress; Stress Classification Model; Deep Belief Network; Machine Learning; KNHANES

MeSH Terms

Classification*
Dataset
Delivery of Health Care
Forests
Korea
Learning
Machine Learning
Methods
Nutrition Surveys
Quality of Life
Sensitivity and Specificity
Support Vector Machine

Figure

• 

  Figure 1 Study design. KNHANES: Korea National Health and Nutrition Examination Survey, DBN: Deep Belief Network.

• 

  Figure 2 Confusion matrix [21].

• 

  Figure 3 Loss function in terms of profiles: (A) profile 1, (B) profile 2, (C) profile 3, and (D) profile 4.

• 

  Figure 4 Deep Belief Network model for stress classification. SBP: systolic blood pressure.

• 

  Figure 5 Sensitivity results. NB: naive Bayesian, DT: decision tree, SVM: support vector machine, DBN: Deep Belief Network.

• 

  Figure 6 Specificity results. NB: naive Bayesian, DT: decision tree, SVM: support vector machine, DBN: Deep Belief Network.

• 

  Figure 7 Accuracy results. NB: naive Bayesian, DT: decision tree, SVM: support vector machine, DBN: Deep Belief Network.

Reference

1. Park JG. Articles on stress and anxiety disorder that destroy me [Internet]. Seoul: Joongdo Daily;2017. cited 2017 Oct 1. Available from: http://www.joongdo.co.kr/jsp/article/article_view.jsp?pq=201704101817.

2. Yoon JH, Lee RZ, Kim MJ. The relationship of self-rated health condition to stress recognition, health related habits, serum biochemical indices, and nutritional intakes in Korean healthy adults. Korean J Food Nutr. 2017; 30(1):83–95.

3. Jeon HG, Sim JM, Lee KC. An empirical analysis of effects of stress on relation between physical activity and health-related quality of life: results from KNHANES 2008 to 2013. J Korea Acad Ind Coop Soc. 2015; 16(8):5351–5363.

4. Hwang JM. Classification of health behaviors of Korean adults using physical activity, smoking, drinking and relationship with mental health. J Korea Soc Wellness. 2016; 11(4):369–379.

5. Lim CY, Kim KH. A Study on the assessment of stress using wireless ECG. J Korea Soc Comput Inf. 2011; 16(2):17–23.

6. Cho YC, Kim MS. Characteristics in HRV (heart rate variability), GSR (galvanic skin response) and skin temperature for stress estimate. J Korea Ind Inf Syst Res. 2015; 20(3):11–18.

7. Xu Q, Nwe TL, Guan C. Cluster-based analysis for personalized stress evaluation using physiological signals. IEEE J Biomed Health Inform. 2015; 19(1):275–281.

8. Sani MM, Norhazman H, Omar HA, Zaini N, Ghani SA. Support vector machine for classification of stress subjects using EEG signals. In : Proceedings of 2014 IEEE Conference on Systems, Process and Control (ICSPC); 2014 Apr 12-14; Kuala Lumpur, Malaysia. p. 127–131.

9. Litjens G, Kooi T, Bejnordi BE, Setio AAA, Ciompi F, Ghafoorian M, et al. A survey on deep learning in medical image analysis. Med Image Anal. 2017; 42:60–88.

10. Hinton GE, Osindero S, Teh YW. A fast learning algorithm for deep belief nets. Neural Comput. 2006; 18(7):1527–1554.

11. Hinton GE, Salakhutdinov RR. Reducing the dimensionality of data with neural networks. Science. 2006; 313(5786):504–507.

12. Lee H, Ekanadham C, Ng AY. Sparse deep belief net model for visual area V2. Adv Neural Inf Process Syst. 2008; 20:873–880.

13. Krizhevsky A, Hinton G. Learning multiple layers of features from tiny images [master's thesis]. Toronto: Department of Computer Science, University of Toronto;2009.

14. Vincent P, Larochelle H, Bengio Y, Manzagol PA. Extracting and composing robust features with denoising autoencoders. In : Proceedings of the 25th International Conference on Machine Learning; 2008 Jul 5-9; Helsinki, Finland. p. 1096–1103.

15. Dahl GE, Yu D, Deng L, Acero A. Context-dependent pre-trained deep neural networks for large vocabulary speech recognition. IEEE Trans Audio Speech Lang Process. 2012; 20(1):30–42.

16. Mishra C, Gupta DL. Deep machine learning and neural networks: an overview. Int J Hybrid Inf Technol. 2016; 9(11):401–414.

17. Abdel-Zaher AM, Eldeib AM. Breast cancer classification using deep belief networks. Expert Syst Appl. 2016; 46:139–144.

18. Tamilselvan P, Wang P. Failure diagnosis using deep belief learning based health state classification. Reliab Eng Syst Saf. 2013; 115:124–135.

19. Fakoor R, Ladhak F, Nazi A, Huber M. Using deep learning to enhance cancer diagnosis and classification. In : Proceedings of the International Conference on Machine Learning (ICML); 2013 Jun 16-21; Atlanta, GA.

20. Korea Center for Disease Control and Prevention. The six Korea National Health & Nutrition Examination Survey 2013-2015 (KNHANES VI) [Internet]. Cheongju: Korea Center for Disease Control and Prevention;c2017. cited 2017 Oct 1. Available from: http://knhanes.cdc.go.kr.

21. Everything About Data Science. Confusion matrix [Internet]. Confusion matrix [Internet]: place unknown: place unknown;c2016. cited 2017 Oct 1. Available from: http://scaryscientist.blogspot.kr/2016/03/confusion-matrix.html.

22. DL4J. Tutorial: deep-belief networks & MNIST [Internet]. place unknown: place unknown: c2017. cited 2017 Oct 1. Available from: https://deeplearning4j.org/deepbeliefnetwork.html.

23. DL4J. A beginner's tutorial for restricted Boltzmann machines [Internet]. place unknown: place unknown: c2017. cited 2017 Oct 1. Available from: https://deeplearning4j.org/restrictedboltzmannmachine.

24. History of deep running: site that summarizes the history of deep running [Internet]. place unknown: Jinseob's Home: c2014. cited 2017 Oct 1. Available from: https://mathemedicine.github.io/deep_learning.html.

25. DL4J. Deep learning for Java [Internet]. place unknown: place unknown: c2017. cited 2017 Oct 1. Available from: https://deeplearning4j.org.

26. Dl4J-examples (Java) [Internet]. place unknown: place unknown: c2017. cited 2017 Oct 1. Available from: https://github.com/deeplearning4j/dl4j-examples.

27. Saito G, Lee BY. Deep learning from scratch. Daejeon: Hanbit Media;2017. p. 103–119. p. 131–143.