Combining Cattle Activity and Progesterone Measurements Using Hidden Semi-Markov Models

  • Jared O’Connell
  • Frede Aakmann Tøgersen
  • Nicolas C. Friggens
  • Peter Løvendahl
  • Søren Højsgaard
Article

Abstract

Hourly pedometer counts and irregularly measured concentration of the hormone progesterone were available for a large number of dairy cattle. A hidden semi-Markov was applied to this bivariate time-series data for the purposes of monitoring the reproductive status of cattle. In particular, the ability to identify oestrus is investigated as this is of great importance to farm management. Progesterone concentration is a more accurate but more expensive method than pedometer counts, and we evaluate the added benefits of a model that includes this variable. The resulting model is biologically sensible, but validation is difficult. We utilize some auxiliary data to demonstrate the model’s performance.

Key Words

Dairy cow EM-algorithm Oestrus detection Online data Streaming data Time series 

Preview

Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.

References

  1. Ball, P. J. H., and Peters, A. R. (2004), Reproduction in Cattle (3rd ed.), Oxford: Blackwell Publishing. CrossRefGoogle Scholar
  2. Baum, L., Petrie, T., Soules, G., and Weiss, N. (1970), “A Maximization Technique Occurring in the Statistical Analysis of Probabilistic Functions of Markov Chains,” The Annals of Mathematical Statistics, 41, 164–171. MathSciNetMATHCrossRefGoogle Scholar
  3. Choi, S., and Wette, R. (1969), “Maximum Likelihood Estimation of the Parameters of the Gamma Distribution and Their Bias,” Technometrics, 11, 683–690. MATHCrossRefGoogle Scholar
  4. Dempster, A., Laird, N., and Rubin, D. (1977), “Maximum Likelihood from Incomplete Data Via the EM Algorithm,” Journal of the Royal Statistical Society. Series B (Methodological), 39, 1–38. MathSciNetMATHGoogle Scholar
  5. Ferguson, J. (1980), Hidden Markov Analysis: An Introduction, Hidden Markov Models for Speech. Google Scholar
  6. Firk, R., Stamer, E., Junge, W., and Krieter, J. (2002), “Automation of Oestrus Detection in Dairy Cows: A Review,” Livestock Production Science, 75, 219–232. CrossRefGoogle Scholar
  7. Forney, G. Jr. (1973), “The Viterbi Algorithm,” Proceedings of the IEEE, 61, 268–278. MathSciNetCrossRefGoogle Scholar
  8. Friggens, N. C., and Løvendahl, P. (2008), “The Potential of On-Farm Fertility Profiles: In-Line Progesterone and Activity Measurements,” in Fertility in Dairy Cows: Bridging the Gaps, British Society of Animal Science, eds. N. C. Friggens, M. D. Royal, R. Smith, Cambridge University Press, Cambridge, pp. 72–78. Google Scholar
  9. Friggens, N. C., Bjerring, M., Ridder, C., Højsgaard, S., and Larsen, T. (2008), “Improved Detection of Reproductive Status in Dairy Cows Using Milk Progesterone Measurements,” Reproduction in Domestic Animals, 43, 113–121. CrossRefGoogle Scholar
  10. Guédon, Y. (2003), “Estimating Hidden Semi-Markov Chains From Discrete Sequences,” Journal of Computational & Graphical Statistics, 12, 604–639. CrossRefGoogle Scholar
  11. Holt, C. (1957), “Forecasting Trends and Seasonals by Exponentially Weighted Moving Averages,” ONR Memorandum, 52. Google Scholar
  12. Hughes, J., Guttorp, P., and Charles, S. (1999), “A Non-Homogeneous Hidden Markov Model for Precipitation Occurrence,” Journal of the Royal Statistical Society (Series C): Applied Statistics, 48, 15–30. MATHCrossRefGoogle Scholar
  13. Jonsson, R., Bjorgvinsson, T., Blanke, M., Poulsen, N., Højsgaard, S., and Munksgaard, L. (2008), “Oestrus Detection in Dairy Cows Using Likelihood Ratio Tests,” The International Federation of Automatic Control, 658–663. Google Scholar
  14. Krogh, A., Mian, I., and Haussler, D. (1994), “A Hidden Markov Model that Finds Genes in E. Coli DNA,” Nucleic Acids Research, 22, 4768–4778. CrossRefGoogle Scholar
  15. Meyer, D. (2002), “Naive Time Series Forecasting Methods,” R News, 2, 7–10. Google Scholar
  16. O’Connell, J., and Højsgaard, S. (2009a, submitted), “Hidden Semi Markov Models for Multiple Observation Sequences—The MHSMM Package for R,” Journal of Statistical Software. Google Scholar
  17. O’Connell, J., and Højsgaard, S. (2009b), “MHSMM: Parameter Estimation and Prediction for Hidden Markov and Semi-Markov Models for Data with Multiple Observation Sequences,” http://cran.r-project.org/web/packages/mhsmm/index.html. R package version 0.3.0.
  18. R Development Core Team (2008), R: A Language and Environment for Statistical Computing, Vienna: R Foundation for Statistical Computing. ISBN 3-900051-07-0. Google Scholar
  19. Rabiner, L. (1989), “A Tutorial on Hidden Markov Models and Selected Applications in Speech Recognition,” Proceedings of the IEEE, 77, 257–286. CrossRefGoogle Scholar
  20. Roelofs, J., van Eerdenburg, F., Soede, N., and Kemp, B. (2005), “Various Behavioral Signs of Estrous and their Relationship with Time of Ovulation in Dairy Cattle,” Theriogenology, 63, 1366–1377. CrossRefGoogle Scholar
  21. Sansom, J., and Thomson, P. (2001), “Fitting Hidden Semi–Markov Models to Breakpoint Rainfall Data,” Journal of Applied Probability, 38, 142–157. MathSciNetCrossRefGoogle Scholar
  22. Winter, P. (1959), “Forecasting Sales by Exponentially Weighted Moving Averages,” Management Science, 6, 324–342. CrossRefGoogle Scholar

Copyright information

© International Biometric Society 2010

Authors and Affiliations

  • Jared O’Connell
    • 1
  • Frede Aakmann Tøgersen
    • 2
  • Nicolas C. Friggens
    • 3
  • Peter Løvendahl
    • 4
  • Søren Højsgaard
    • 4
  1. 1.Wellcome Trust Centre for Human GeneticsUniversity of OxfordOxfordUK
  2. 2.Modeling, Statistics and Risk AnalysisVestas R&DAlsvejDenmark
  3. 3.INRA UMR 791 Modélisation Systémique Appliquée aux Ruminants (MoSAR)AgroParisTechParis CedexFrance
  4. 4.Department of Genetics and BiotechnologyAarhus UniversityÅrhusDenmark

Personalised recommendations