Pattern Recognition in Intensive Care Online Monitoring

  • R. Fried
  • U. Gather
  • M. Imhoff
Part of the Studies in Fuzziness and Soft Computing book series (STUDFUZZ, volume 96)


Clinical information systems can record numerous variables describing the patient’s state at high sampling frequencies. Intelligent alarm systems and suitable bedside decision support are needed to cope with this flood of information. A basic task here is the fast and correct detection of important patterns of change such as level shifts and trends in the data. We present approaches for automated pattern detection in online-monitoring data. Several methods based on curve fitting and statistical time series analysis are described. Median filtering can be used as a preliminary step to reduce the noise and to remove clinically irrelevant short term fluctuations.


Control Chart Forecast Error Median Filter Level Shift Dynamic Linear Model 
These keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.


Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.


  1. [1]
    Mäkivirta, A. (1989), “Towards Reliable and Intelligent Alarms by Using Median Filters,” Research Report 660, Technical Research Centre of Finland.Google Scholar
  2. [2]
    Morris, A. and Gardner, R. (1992), “Computer applications,” in: Hall, J., Schmidt, G., and Wood, L. (Eds.), Principles of Critical Care, McGraw Hill, New York, pp. 500–514.Google Scholar
  3. [3]
    Miller, G. (1956), “The marginal number seven, plus or minus two: some limits to our capacity for processing information,” Psychol. Rev., vol. 63, pp. 81–97.PubMedCrossRefGoogle Scholar
  4. [4]
    Jennings, D., Amabile, T., and Ross, L. (1982), “Informal covariation asessments: data-based versus theory-based judgements,” in: Kahnemann, D., Slovic, P., and Tversky, A. (eds.), Judgment under Uncertainty: Heuristics and Biases, Cambridge University Press, Cambridge, pp. 211–230.CrossRefGoogle Scholar
  5. [5]
    Guyatt, G., Drummond, M., Feeny, D., Tugwell, P., Stoddart, G., Haynes, R., Bennett, K., and LaBelle, R. (1986), “Guidelines for the clinical and economic evaluation of health care technologies,” Soc. Sci. Med., vol. 22, pp. 393–408.PubMedCrossRefGoogle Scholar
  6. [6]
    Morris, A. (1998), “Algorithm-based decision making,” in: Tobin, J.A. (Ed.), Principles and practice of intensive care monitoring, McGraw Hill, New York, pp. 1355–1381.Google Scholar
  7. [7]
    Lawless, S.T. (1994), “Crying wolf: false alarms in a pediatric intensive care unit,” Critical Care Medicine, vol. 22, pp. 981–985.PubMedCrossRefGoogle Scholar
  8. [8]
    Wiklund, L., Hök, B., Stähl, K., and Jordeby-Jönsson, A. (1994), “Postanaesthesia monitoring revisited: frequency of true and false alarms from different monitoring devices,” J. Clin. Anesth., vol. 6, pp. 182–188.PubMedCrossRefGoogle Scholar
  9. [9]
    O’Carrol, T. (1986), “Survey of alarms in an intensive therapy unit,” Anesthesia, vol. 41, pp. 742–744.CrossRefGoogle Scholar
  10. [10]
    Haimowitz, I.J., Le, P.P., and Kohane, I.S. (1995), “Clinical monitoring using regression-based trend templates,”Art. Intel. Med., vol. 7, pp. 473–496.Google Scholar
  11. [11]
    Haimowitz, I.J. and Kohane, I.S. (1996), “Managing temporal worlds for medical trend diagnosis,” Art. Intel. Med., vol. 8, pp. 299–321.CrossRefGoogle Scholar
  12. [12]
    Miksch, S., Horn, W., Popow, C., and Paky, F. (1996), “Utilizing temporal data abstraction for data validation and therapy planning for artificially ventilated newborn infants,” Art. Intel. Med., vol. 8, pp. 543–576.CrossRefGoogle Scholar
  13. [13]
    Smith, A.F.M. and West, M. (1983), “Monitoring renal transplants: an application of the multiprocess Kalman filter,” Biometrics, vol. 39, pp. 867–878.PubMedCrossRefGoogle Scholar
  14. [14]
    Gather, U., Fried, R., Lanius, V., and Imhoff, M. (2001), “Online monitoring of high-dimensional physiological time series — a case-study,” Estadistica. (To appear.)Google Scholar
  15. [15]
    Shahar, Y. and Musen, M.A. (1993), “RESUME: a temporal-abstraction system for patient monitoring,” Computers and Biomedical Research, vol. 26, pp. 255–273.PubMedCrossRefGoogle Scholar
  16. [16]
    Shahar, Y. and Musen, M.A. (1996), “Knowledge-based temporal abstraction in clinical domains,” Art. Intel. Med., vol. 8, pp. 267298.Google Scholar
  17. [17]
    Mäkivirta, A., Koski, E., Kari, A., and Sukuvaara, T. (1991), “The median filter as a preprocessor for a patient monitor limit alarm system in intensive care,” Computer Methods and Programs in Medicine, vol. 34, pp. 134–149.Google Scholar
  18. [18]
    Justusson, B.I. (1978), “Noise reduction by median filtering,” Pro. 4th Int. Joint Conf Pattern recognition, Kyoto, Japan, Nov. 1978, pp. 502–504.Google Scholar
  19. [19]
    Justusson, B.I. (1981), “Median filtering: statistical properties,” in: Huang, T.S. (Ed.), Topics in Applied Physics, Two-Dimensional Signal Processing II, Springer-Verlag, Berlin.Google Scholar
  20. [20]
    Nieminen, A., Neuvo, Y., and Mitra, U. (1988), “Algorithms for real-time trend detection,” ICASSP International Conference on Acoust., Speech and Signal Proc. 1988, IEEE, New York, pp. 1530–1532.CrossRefGoogle Scholar
  21. [21]
    Nieminen, A., Neuvo, Y., and Mitra, U. (1989), “Algorithms for real-time trend detection,” Signal processing, vol. 18, pp. 1–15.CrossRefGoogle Scholar
  22. [22]
    Endresen, J. and Hill, D.W. (1977), “The present state of trend detection and prediction in patient monitoring,” Intensive Care Medicine, vol. 3, pp. 15–26.PubMedCrossRefGoogle Scholar
  23. [23]
    Woodward, W.A. and Gray, H.L. (1993), “Global warming and the problem of testing for trend in time series data, ” Journal of Climate, vol. 6, pp. 953–962.CrossRefGoogle Scholar
  24. [24]
    Hill, D.W. and Endresen, J. (1978), “Trend recording and forecasting in intensive care therapy,” British Journal of Clinical Equipment, January, pp. 5–14.Google Scholar
  25. [25]
    Imhoff, M. and Bauer, M. (1996), “Time series analysis in critical care monitoring,” New Horizons, vol. 4, pp. 519–531.PubMedGoogle Scholar
  26. [26]
    Gordon, K. and Smith, A.S.M. (1990), “Modeling and monitoring biomedical time series,” J Americ. Statist. Assoc., vol. 85, pp. 328337.Google Scholar
  27. [27]
    Hepworth, J.T., Handrickson, S.G., and Lopez, J. (1994), “Time series analysis of physiological response during ICU visitation,” West J. Nurs. Res., vol. 16, pp. 704–717.PubMedCrossRefGoogle Scholar
  28. [28]
    Imhoff, M., Bauer, M., Gather, U., and Löhlein, D. (1998), “Statistical pattern detection in univariate time series of intensive care online monitoring data,” Intensive Care Medicine, vol. 24, pp. 13051314.Google Scholar
  29. [29]
    Bauer, M., Gather, U., and Imhoff, M. (1999), “The Identification of Multiple Outliers in Online Monitoring Data,” Technical Report 29/1999, SFB 475, Department of Statistics, University of Dortmund, Germany.Google Scholar
  30. [30]
    Brockwell, P.J. and Davis, R.A. (1987), Time Series: Theory and Methods, 2nd ed., Springer, New York.CrossRefGoogle Scholar
  31. [31]
    Box, G.E.P., Jenkins, G.M., and Reinsel, G.C. (1994), Time Series Analysis. Forecasting and Control, 3rd ed., Prentice-Hall, Englewood Cliffs.Google Scholar
  32. [32]
    Fox, A.J. (1972), “Outliers in time series,” J Roy. Statist. Soc. Ser. B, vol. 34, pp. 350–363.Google Scholar
  33. [33]
    Chang, I., Tiao, G.C., and Chen, C. (1988), “Estimation of time series parameters in the presence of outliers,” Technometrics, vol. 30, pp. 193–204.CrossRefGoogle Scholar
  34. [34]
    Tsay, R.S. (1988), “Outliers, level shifts and variance changes in time series,” J Forecasting, vol. 7, pp. 1–20.CrossRefGoogle Scholar
  35. [35]
    Chen, C., and Liu, L. (1993), “Joint estimation of model parameters and outlier effects in time series,” J Am. Stat. Assoc., vol. 88, pp. 284–297.Google Scholar
  36. [36]
    Pena, D., Tiao, G.C., and Tsay, R.S. (2000), A Course in Time Series Analysis, Wiley, New York.CrossRefGoogle Scholar
  37. [37]
    Muirhead, C.R. (1986), “Distinguishing outlier types in time series,” J. R. Statist. Soc. Ser. B, vol. 48, pp. 39–47.Google Scholar
  38. [38]
    Ljung, G.M. (1993), “On outlier detection in time series,” J. Roy. Stat. Soc. B, vol. 55, pp. 559–567.Google Scholar
  39. [39]
    Imhoff, M., Bauer, M., Gather, U., and Löhlein, D. (1997), “Time series analysis in intensive care medicine,” Applied Cardiopulmonary Pathophysiology, vol. 6, pp. 263–281.Google Scholar
  40. [40]
    Johnson, R.A. and Bagshaw, M. (1974), “The effects of serial correlation on the performance of the CUSUM tests,” Technometrics, vol. 16, pp. 103–122.CrossRefGoogle Scholar
  41. [41]
    Padgett, C.S., Thombs, L.A., and Padgett, W.J. (1992), “On the a-risks for Shewhart control charts,” Communications in Statistics–Simulation and Computation, vol. 21, pp. 1125–1147.CrossRefGoogle Scholar
  42. [42]
    Lu, C.-W. and Reynolds, M.R. Jr. (1999), “EWMA control charts for monitoring the mean of autocorrelated processes,” Journal of Quality Technology, vol. 31, pp. 166–187.Google Scholar
  43. [43]
    Kramer, H. and Schmid, W. (2000), “The influence of parameter estimation on the ARL of Shewhart type charts for time series,” Statistical Papers, vol. 41, pp. 173–196.CrossRefGoogle Scholar
  44. [44]
    Tseng, S. and Adams, B.M. (1994), “Monitoring autocorrelated processes with an exponentially weighted moving average forecast,” J. Statist. Comp. Simul., vol. 50, pp. 187–195.CrossRefGoogle Scholar
  45. [45]
    Adams, B.M. and Tseng, I.-T. (1998), “Robustness of forecast-based monitoring schemes,” Journal of Quality Technology, vol. 30, pp. 328–339.Google Scholar
  46. [46]
    Runger, G.C. and Willeman, T.R. (1995), “Model-based and model-free control of autocorrelated processes,” Journal of Quality Technology, vol. 27, pp. 283–292.Google Scholar
  47. [47]
    Hembree, G.B. (1994), “Recursive estimation of the weighting factor for EWMA control charts from autocorrelated data,” ASAProQIPr, 39–43.Google Scholar
  48. [48]
    Lin, W.S.W. and Adams, B.M. (1996), “Combined control charts for forecast-based monitoring schemes,” Journal of Quality Technology, vol. 28, pp. 289–301.Google Scholar
  49. [49]
    Zhang, N.F. (1998), “A statistical control chart for stationary process data,” Technometrics, vol. 40, pp. 24–38.CrossRefGoogle Scholar
  50. [50]
    Montgomery, D.C. and Mastrangelo, C.M. (1991), “Some statistical process control methods for autocorrelated data,” Journal of Quality Technology, vol. 23, pp. 179–193.Google Scholar
  51. [51]
    Timmer, D.H., Pignatiello, J. Jr., and Longnecker, M. (1998), “The development and evaluation of CUSUM-based control charts for an AR(1) process,” IEEE Transactions, vol. 30, pp. 525–534.Google Scholar
  52. [52]
    Lambert, C.R., Raymenants, E., and Pepine, C.J. (1995), “Time-series analysis of long-term ambulatory myocardial ischemia: effects of beta-adrenergic and calcium channel blockade,” Am. Heart J, vol. 129, pp. 677–684.PubMedCrossRefGoogle Scholar
  53. [53]
    Schack, B. and Grieszbach, G. (1994), “Adaptive methods of trend detection and their applications in analysing biosignals,” Biometrical Journal, vol. 36, pp. 429–452.CrossRefGoogle Scholar
  54. [54]
    Trigg, D.W. (1964), “Monitoring a forecasting system,” Operational Research Quarterly, vol. 15, pp. 271–274.CrossRefGoogle Scholar
  55. [55]
    Gardner, E.S. (1983), “Automatic monitoring of forecast errors,” Journal of Forecasting, vol. 2, pp. 1–21.CrossRefGoogle Scholar
  56. [56]
    Kennedy, R.R. (1995), “A modified Trigg’s tracking variable as an `advisory alarm’ during anaesthesia,” International Journal of Clinical Monitoring and Computing, vol. 12, pp. 197–204.PubMedCrossRefGoogle Scholar
  57. [57]
    Stoodley, K.D.C. and Mirnia, M. (1979), “The automatic detection of transients, step changes and slope changes in the monitoring of medical time series,” The Statistician, vol. 28, pp. 163–170.CrossRefGoogle Scholar
  58. [58]
    Högel, J. (2000), “Applications of statistical process control techniques in medical fields,” Allg. Stat. Archiv, vol. 84, pp. 337–359.CrossRefGoogle Scholar
  59. [59]
    West, M. and Harrison, J. (1989), Bayesian Forecasting and Dynamic Models, Springer, New York.CrossRefGoogle Scholar
  60. [60]
    Wasserman, G.S. and Sudjianto, A. (1993), “Short run SPC based upon the second order dynamic linear model for trend detection,” Communications in Statistics–Computation and Simulation, vol. 22, 1011–1036.CrossRefGoogle Scholar
  61. [61]
    Pena, D. (1990), “Influential observations in time series,” I Business & Economic Statistics, vol. 8, pp. 235–241.Google Scholar
  62. [62]
    De Jong, P., and Penzer, J. (1998), “Diagnosing shocks in time series,” J. Americ. Statist. Assoc., vol. 93, pp. 796–806.CrossRefGoogle Scholar
  63. [63]
    Gather, U., Fried, R., and Imhoff, M. (2000), “Online classification of states in intensive care,” in: Gaul, W., Opitz, O., and Schader, M. (Eds.), Festschrift in Honor to Hans-Hermann Bock’s 60th Birthday, Data Analysis, Classification, and Applications, Springer, Berlin, pp. 413–428.Google Scholar
  64. [64]
    Becker, C. and Gather, U. (2000), “The masking breakdown point of multivariate outlier identification rules,” I Americ. Statist. Assoc., vol. 94, pp. 947–955.CrossRefGoogle Scholar
  65. [65]
    Fried, R., Gather, U., and Imhoff, M. (2001), “The online detection of a monotone trend in a time series,” Preprint, Department of Statistics, University of Dortmund, Germany.Google Scholar
  66. [66]
    Abelson, R.P. and Tukey, J.W. (1963), “Efficient utilization of non-numerical information in quantitative analysis: general theory and the case of simple order,” Ann. Math. Statist., vol. 34, pp. 13471369.Google Scholar
  67. [67]
    Brillinger, D.R. (1989), “Consistent detection of a monotonic trend superposed by a stationary time series,” Biometrika, vol. 76, pp. 23–30.CrossRefGoogle Scholar
  68. [68]
    Härdle, W. (1990), Applied Nonparametric Regression, Cambridge University Press, Cambridge.CrossRefGoogle Scholar
  69. [69]
    Morik, K., Imhoff, M., Brockhausen, P., Joachims, T., and Gather, U. (2000), “Knowledge discovery and knowledge validation in intensive care,” Art. Int. Med., vol. 19, pp. 225–249.CrossRefGoogle Scholar
  70. [70]
    Friedmann, J.H. (1994), “An overview of predictive learning and function approximation,” in: Cherkassky, V., Friedmann, J.H., and Wechsler, H. (Eds.), From Statistics to Neural Networks, Springer, Berlin et al., pp. 1–61.CrossRefGoogle Scholar
  71. [71]
    Fried, R., Gather, U., and Imhoff, M. (2000), “Some statistical methods in intensive care online monitoring — a review,” in: Brause, R.W. and Hanisch, E. (Eds.), Medical Data Analysis, Springer, Berlin, pp. 67–77.CrossRefGoogle Scholar
  72. [72]
    Gather, U., Imhoff, M., and Fried, R. (2000), “Graphical Models for Multivariate Time Series from Intensive Care Monitoring,” Technical Report 33/2000, SFB 475, University of Dortmund, 44221 Dortmund, Germany.Google Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 2002

Authors and Affiliations

  • R. Fried
  • U. Gather
  • M. Imhoff

There are no affiliations available

Personalised recommendations