DOI QR코드

DOI QR Code

Health State Clustering and Prediction Based on Bayesian HMM

Bayesian HMM 기반의 건강 상태 분류 및 예측

  • 신봉기 (부경대학교 IT융합응용공학과)
  • Received : 2017.02.06
  • Accepted : 2017.07.21
  • Published : 2017.10.15

Abstract

In this paper a Bayesian modeling and duration-based prediction method is proposed for health clinic time series data using the Hierarchical Dirichlet Process Hidden Markov Model (HDP-HMM). HDP-HMM is a Bayesian extension of HMM which can find the optimal number of health states, a number which is highly uncertain and even difficult to estimate under the context of health dynamics. Test results of HDP-HMM using simulated data and real health clinic data have shown interesting modeling behaviors and promising prediction performance over the span of up to five years. The future of health change is uncertain and its prediction is inherently difficult, but experimental results on health clinic data suggests that practical long-term prediction is possible and can be made useful if we present multiple hypotheses given dynamic contexts as defined by HMM states.

본 논문은 계층적 디리슐레 과정(HDP)과 은닉 마르코프 모형(HMM)이 결합된 베이스 통계학적 방법과 HMM의 상태 지속 정보를 이용한 건강 상태 예측 방법을 제안한다. HDP-HMM은 베이스 방법의 HMM 확장 모형으로서 건강의 동적 특성을 고려하여 불확실하고 가늠하기조차도 어려운 건강 상태의 수를 추정할 수 있게 해준다. 모의 데이터와 실제 건건 검진 데이터를 이용한 시험을 통하여 흥미 있는 행동 특성을 볼 수 있었으며 최대 5년까지로 제한한 미래 예측도 충분한 가능함을 확인하였다. 미래는 불확실하며 예측 문제는 본질적으로 어렵다. 그러나 본 연구의 실험 결과로 동적인 문맥 하에서 다중 후보 가설을 제시함으로서 실용 가능한 건강상태의 장기 예측이 가능하다는 것을 읽을 수 있었다.

Keywords

References

  1. Y.-W. Kim, B.-K. Sin, and H. Choi, "HMM-based health state clustering and Prediction," Proc. 5th Int. Conf. on Ubiq. Computing App. and Wireless Sensor Netowrk (UCAWSN-16), 2016.
  2. R. Y. Coley, et al., "A Bayesian hierarchical model for prediction of latent health states from multiple data sources with application to active surveillance of prostrate cancer," [Online]. Available: http://arXive eprint arXiv:1508.07511 [stat.ME], Aug. 2015.
  3. L. R. Rabiner, "A tutorial on hidden Markov models and selected applications in speech recognitions," Proc. IEEE, Vol. 77, pp. 257-286, 1989. https://doi.org/10.1109/5.18626
  4. M. Beal, Variational algorithms for approximate Bayesian inference, PhD thesis, Univ. London, 2003.
  5. P. I. Green, "Reversible jump Markov chain Monte Carlo computation and Bayesian model determination," Biometrika, Vol. 82, pp. 711-732, 1995. https://doi.org/10.1093/biomet/82.4.711
  6. M. Beal, Z. Ghahramani, and C. Rasmussen, "The infinite hidden Markov model," NIPS, 14, 2002.
  7. G.H. Juang and L.R. Rabiner "Mixture autoregressive hidden Markov models for speech signals," IEEE Trans. ASSP, Vol. ASSP-33, No. 6, pp. 1404-1413, 1985.
  8. M.J. Russell, and R.K. Moore, "Explicit modeling of state occupancy in hidden Markov models for automatic speech recognition," Proc. ICASSP'85 (Tampa, FL), pp. 5-8, 1985.
  9. B.-K. Sin, and J.H. Kim, "Nonstationary hidden Markov model," Signal Porcessing, Vol. 46, No. 1, pp. 31-46, Sep. 1995. https://doi.org/10.1016/0165-1684(95)00070-T
  10. B.-K. Sin, "Gamma CDF-based HMM state duration modeling," Journal of KIISE: SW and Appl., Vol. 40, No. 12, pp. 757-763, 2013.
  11. Z. Ghahramani, and M. Jordan, "Factorial hidden Markov models," Machine Learning, Vol. 29, pp. 245-273, 1997. https://doi.org/10.1023/A:1007425814087
  12. Y.W. Teh, M. Jordan, M. Beal, and D. Blei, "Hierarchical Dirichlet processes," J. Amer .Stat. Assoc., Vol. 101, pp. 1566-1581, 2006. https://doi.org/10.1198/016214506000000302
  13. T. Ferguson, "A Bayesian analysis of some nonparametric problems," Annals of Statistics, Vol. 1, No. 2, pp. 209-230, Mar. 1973. https://doi.org/10.1214/aos/1176342360
  14. J. Sethuraman, "A constructive definition of Dirichlet priors," Statistica Sinica, Vol. 4, pp. 639-650, 1994.
  15. C.M. Bishop, Pattern Recognition and Machine Learning, Springer, 2006.
  16. S. Geman and D. Geman, "Stochastic relaxation, Gibbs distributions, and the Bayesian restoration of images," IEEE Tr. PAMI, Vol. 6, pp. 721-741, 1984.
  17. H. Robbins, "An empirical Bayes approach to statistics," Proc. 3rd Berkeley Symp. on Math. Stat. and Prob., Vol. 1, pp. 157-163, 1956.
  18. National Health Insurance Sharing Service, Korea, [Online]. Available: http://nhiss.nhis.or.kr/op/it/index.do, (Accessed in December, 2016)
  19. R.J. Little and B.B. Rubin, Statistical Analysis with Missing Data, John Wiley & Sons, 2014.
  20. Y. Zhang, Prediction of financial time series with hidden Markov models, Masters thesis, Simon Fraser University, 2004.
  21. M. Stanke, Gene prediction with a hidden Markov model, Doctoral dissertation, Georg-August Universitat, 2003.
  22. S.M. Ross, Introduction to Probability Models, (11th ed.), p. 187, 2014.