Abstract
The use of a multivariate statistical technique, canonical analysis, for the assignment of patients to diagnostic categories is presented. The classification of thyroid status using clinical findings is investigated to exemplify the application of the technique. Emphasis is placed on the stages in the data analysis that are generic in examining any set of medical data for the purpose of automatically classifying patients.
Similar content being viewed by others
References
Lezotte, D. C., A multivariate laboratory data analysis system: Introduction.J. Med. Systems, 1:293–298, 1977.
Adelstein, S. J., Parker, R., and Wagner, H. N., First phase in objective evaluation of new diagnostic tests.Inv. Rad. 5:153–163, 1970.
Rao, C. R.,Linear statistical inference and its application, Wiley, New York, 1967.
Fisher, R. A. The use of multiple measurements in taxonomic problems.Am. Eug. (London) 7:179, 1936.
Bartlett, M. S.,Multivariate Statistics. Theoretical and Mathematical Biology, H. J. Morowitz and T. E. Waterman, (Eds.), Academic Press, New York, 1965.
Gorry, G. A., and Barnett, G. O. Sequential diagnosis by computer.J. A. M. A. 205:849, 1968.
Lincoln, T. L., and Parker, R. D., Medical diagnosis using Bayes Theorem.Health Serv. Res. 2:34–45, 1967.
Parker, R. D., and Boyd, J. L., A comparison of a discriminant versus a clustering analysis of a patient classification for chronic disease care.Med. Care 12:944–957, 1974.
Parker, R. D., Yamamoto, L. A., and Miller, W. R., Interaction effects analysis of platelet transfusion data.Transfusion 14:567–573, 1974.
Yockey, H. P., Platzman, R. L., and Quastler, H.,Symposium on Information Theory in Biology, Pergamon Press, Elmsford, New york, 1956, pp. 3–48.
Author information
Authors and Affiliations
Rights and permissions
About this article
Cite this article
Parker, R.D. An illustration of the technique of canonical analysis. J Med Syst 2, 161–170 (1978). https://doi.org/10.1007/BF02222316
Issue Date:
DOI: https://doi.org/10.1007/BF02222316