Abstract
The paper describes an approach to intelligent ischaemia event detection based on ECG ST-T segment analysis. ST-T trends are processed by means of a Bayesian forecasting approach using the multistate Kalman filter. A complete procedure, intended for use in CCU/ICU monitoring areas, is proposed, in order to give the clinician an intelligent monitoring tool. The approach serves to describe trends and their changes in a symbolic way. A novel aspect is its ability to observe certain features of ST-T elevation/depression not detected by other means, and to reject artefacts and erroneous events. A sensivity of 89.58% and a predictivity of 84.31% are obtained on selected records of the European ST-T database. Using a restriction on event amplitude, the predictivity is raised to 95.55%. An ischaemia sensitivity index of 1·2 was determined. The method has been shown to be a robust and practical trend analysis tool, and seems to be appropriate for numeric/symbolic transformations in next-generation intelligent monitoring systems.
This is a preview of subscription content, log in via an institution to check access.
Similar content being viewed by others
References
Avent, R. K., andCharlton, J. D. (1990): ‘A critical review of trend-detection methodologies for biomedical monitoring systems,’Crit. Rev. Biomed. Eng.,17, pp. 621–659
Clark, K., McLear, P., Kortas, R., Mead, C., andThomas, Jr., L. (1981): ‘ARGUS/2H detection of ST-segment changes in ambulatory ECG recordings,’Comput. Cardiol., pp. 27–31
De Lucia, F., Pasaariello, G., Villegas, G., andMora, F. (1990): ‘Microcomputer based cornary care unit central station,’J. Clin. Eng.,15, pp. 381–389
Elliot, C. D. (1991): ‘A high resolution electrocardiogram ST segment trend generator.’ MSc thesis, The University of Oklahoma, Graduate College, Norman-Oklahoma, USA
Factor, M., Gelernter, D. H., Kolb, C. E., Miller, P. L., andSittig, D. F. (1991): ‘Real-time data fusion in the intensive care unit,’Comput.,24, (12), pp. 45–54
Gordon, K. (1986): ‘The multi-state Kalman filter in medical monitoring,’Comput. Method Prog. Biomed.,23, pp. 147–154
Gritzali, F. (1988): ‘Towards a generalised scheme for QRS detection in ECG waveforms,’Signal Process.,15, pp. 183–192
Harrison, P. J., andStevens, C. F. (1971): ‘A Bayesian approach to short-term forecasting,’Oper. Res. Q.,22, pp. 341–362
Harrison, P. J., andStevens, C. F. (1976): ‘Bayesian forecasting’, with discussion,R. Stat. Soc., B,38, pp. 205–247
Jager, F., Moody, G. B., Taddei, A., andMark, R. G. (1992a): ‘Performance measures for algorithms to detect transientichaemic ST segment changes,’Comput Cardiol. (1991), pp. 369–372
Jager, F., Mark, R. G. andMoody, G. B. (1992b): ‘Analysis of transient ischaemic ST segment changes during ambulatory monitoring,’Comput. Cardiol. (1991), pp. 453–456
Kalman, R. E. (1960): ‘A new approach to linear filtering and prediction problems,’J. Basic Eng.,82, p. 35
Krucoff, M. W. (1989): ‘Electrocardiographic monitoring and coronary occlusion,’J. Electrocardiol.,22, supplement, pp. 232–237
Krucoff, M. W. (1990): ‘Comprehensive ischemia monitoring in unstable coronary syndromes,’Coronary Acute Care,1, (2), pp. 2–10
Mora, F., Passariello, G., Carrault, G., andLe Pichon, J. P. (1993): ‘Intelligent patient monitoring and management systems: a review,’IEEE Eng. Med. Biol. Mag.,12, (4), pp. 23–33
Oates, J., Cellar, B., Bernstein, B. P., andFreeman, S. B. (1989): ‘Real-time detection of ischaemic ECG changes using quasiorthogonal leads and artificial intelligence,’Comput. Cardiol. (1988), pp. 89–92
Presedo, J., Barro, S., Palacios, F., Ruiz, R., Vila, J., andBugarin, A. (1992): ‘Detection of significant variations of the ST segment for the determination of ischaemic episodes: Proc. IEEE-EMBS Ann. Conf., Paris, France, pp. 540–541
Shook, T. L., Valvo, V., Hubelbank, M., Feldman, C. L., andStone, P. H. (1988): ‘Varidation of a new algorithm for detection and quantification of ischaemic ST segment changes during ambulatory electrocardiography,’Comput. Cardiol. (1987), pp. 57–62
Sittig, D., andFactor, M. (1990): ‘Physiologic trend detection and artifact rejection: a parallel implementation of multi-state Kalman filtering algorithm,’Comput. Method Prog. Biomed.,31, pp. 1–10
Smith, A. F. M. andWest, M. (1983): ‘Monitoring renal transplants: an application of the multi-process Kalman filter,’Biometrics,39, pp. 867–878
Stamkopoulos, T., Strinzis, M., Pappas, C., andMaglaveras, N. (1992): ‘One-lead ischemia detection using a new back-propagation algorithm and the European ST-T Database,’Comput. Cardiol. (1992), pp. 663–666
Suzuki, Y., andOno, K. (1992): ‘Personal computer system for ECG ST-segment recognition based on neural networks,’Med. Biol. Eng. Comput.,30, (1), pp. 2–8
Taddei, A., Biagini, A., Distante, G., Marchesi, C., Mazzei, M. G., Pisani, P., Roggero, N., andZeelenberg, C. (1990): ‘An annotated database aimed at performance evaluation of algorithms for ST-T change analysis,’Comput. Cardiol. (1989), pp. 117–120
Weisner, S., Tompkins, W., andTompkins, B. (1982): ‘A compact, microprocessor based ECG ST-segment analyzer for the operating room,’IEEE Trans. BME-29, pp. 642–649
Author information
Authors and Affiliations
Rights and permissions
About this article
Cite this article
Bosnjak, A., Bevilacqua, G., Passariello, G. et al. An approach to intelligent ischaemia monitoring. Med. Biol. Eng. Comput. 33, 749–756 (1995). https://doi.org/10.1007/BF02523005
Received:
Accepted:
Issue Date:
DOI: https://doi.org/10.1007/BF02523005