Fault Diagnostic of Variance Shifts in Clinical Monitoring Using an Artificial Neural Network Input Gain Measurement Approximation (ANNIGMA)

  • Nadeera Gnan Tilshan GunaratneEmail author
  • Mali Abdollahian
  • Shamsul Huda
Conference paper
Part of the Advances in Intelligent Systems and Computing book series (AISC, volume 738)


Condition of a patient in an intensive care unit is assessed by monitoring multiple correlated variables with individual observations. Individual monitoring of variables leads to misdiagnosis. Therefore, variability of the correlated variables needs to be monitored simultaneously by deploying a multivariate control chart. Once the shift from the accepted range is detected, it is vital to identify the variables that are responsible for the variance shift detected by the chart. This will aid the medical practitioners to take the appropriate medical intervention to adjust the condition of the patient. In this paper, Multivariate Exponentially Weighted Moving Variance chart has been used as the variance shift identifier. Once the shift is detected, authors for the first time have used ANNIGMA to identify the variables responsible for variance shifts in the condition of the patient and rank the responsible variables in terms of the percentage of their contribution to the variance shift. The performance of the proposed ANNIGMA has been measured by computing average classification accuracy. A case study based on real data collected from ICU unit shows that ANNIGMA not only improve the diagnosis but also speed up the variable identification for the purpose of appropriate medical diagnosis.


MEWMV chart Neural networks Clinical monitoring Univariate moving range char Multivariate variability 


  1. 1.
    L. Huwang, A.B. Yeh, C. Wu, Monitoring multivariate process variabilty for individual observations. J. Qual. Technol. 39(3), 258–278 (2007)CrossRefGoogle Scholar
  2. 2.
    N.G.T. Gunaratne, M. Abdollahian, S. Huda, Monitoring multivariate progress variability after heart surgery, in Innovative Trends in Multidisciplinary Academic Research, Kuala Lumpur, 2014Google Scholar
  3. 3.
    C. Cheng, H. Cheng, Identifying the sources of variance shifts in the multivariate process using neural networks and support vector machines. Expert. Syst. Appl. 35(1), 198–206 (2008)CrossRefGoogle Scholar
  4. 4.
    S.T.A. Niaki, B. Abbasi, Fault diagnosis in multivariate control charts using artificial nueral networks. Qual. Reliab. Eng. 21(8), 825–840 (2005)CrossRefGoogle Scholar
  5. 5.
    C. lOW, C. Hsu, F. Yu, Analysis of variations in a multi-variate process using neural network. Int. J. Adv. Manuf. Technol. 22(11), 911–121 (2003)CrossRefGoogle Scholar
  6. 6.
    M.R. Maleki, A. Amiri, S.M. Mousavi, Step change point estimation in the multivarite-attribute process variabilty using artificial neural networks and maximum likelihood estimation. J. Ind. Eng. Int. 11(4), 505–515 (2015)CrossRefGoogle Scholar
  7. 7.
    S.X. Yin, H.R. Karimi, X. Zhu, Study on support vector machine based faulty detection in tennessee eastman process. Abstr. Appl. Anal. 2014, 1–8 (2014)Google Scholar
  8. 8.
    S.R. Gunn, Support vector machines for classification and regression, Faculty of Engineering, Science and Mathematics, School of Electronics and Computer Science. 1–66 (1998)Google Scholar
  9. 9.
    R. Malhotra, Comparative analysis of statistical and machine learning methods for predicting faulty modules. Appl. Soft Comput. 21, 286–297 (2014)CrossRefGoogle Scholar
  10. 10.
    V. Venkatasubramanian, R. Rengaswamy, S.N. Kavuri, K. Yin, A review of process fault detection and diagnosis part III: process history based methods. Comput. Chem. Eng. 27(3), 327–346 (2003) CrossRefGoogle Scholar
  11. 11.
    S. Du, J. Lv, L. Xi, On-line classifying process mean shifts in multivariate control charts based on multicalss support vector machines. Int. J. Prod. Res. 50(22), 6288–6310 (2012)CrossRefGoogle Scholar
  12. 12.
    S. Huda, M. Abdollahian, M. Mammadav, J. Yearwood, S. Ahmed, I. Sultan, A hybrid wrapper-filter approach to detect the source(s) of out-of-control signals in multivariate manufacturing process. Eur. J. Oper. Res. 237(3), 857–870 (2014)CrossRefGoogle Scholar
  13. 13.
    C. Hsu, H. Huang, D. Schuschel, The ANNIGMA-Wrapper approach to fast feature selection for neural nets. IEEE Trans. Syst. Man Cybern. B Cybern. 32(2), 207–212 (2002)CrossRefGoogle Scholar
  14. 14.
    N.G.T. Gunaratne, M.A. Abdollahian, S. Huda, J. Yearwood, Exponentially weighted control charts to monitor multivariate process variability for high dimensions. Int. J. Prod. Res. 55(17), 4948–4962 (2017)CrossRefGoogle Scholar
  15. 15.
    J.S. Rosenthal, Parallel computing and Monte carlo algorithms. Far East J. Theor. Stat. 4, 207–236 (2000)MathSciNetzbMATHGoogle Scholar

Copyright information

© Springer International Publishing AG, part of Springer Nature 2018

Authors and Affiliations

  • Nadeera Gnan Tilshan Gunaratne
    • 1
    Email author
  • Mali Abdollahian
    • 1
  • Shamsul Huda
    • 2
  1. 1.School of Science, RMIT UniversityMelbourneAustralia
  2. 2.School of Information Technology, Deakin UniversityMelbourneAustralia

Personalised recommendations