Classification of New Cases: Some Aspects of Single and Multivariate Analysis

  • J. P. A. Baak
  • F. A. Langley
  • J. Hermans


In the preceding chapters we have discussed the advantages of quantitation, the factors which influence the subjective diagnostic process and the implicit or subconscious presence of a structural model in the qualitative or quantitative analysis of a structure. Chapter 3 and sections A.2.1 and A.2.2 deal with the collection of quantitative data. In the present chapter, some aspects of techniques used for the discrimination of groups of patients on the basis of quantitative data will be discussed. Moreover, the classification of new patients will be considered. In agreement with the aims of this manual, and as we hope, also in accordance with the desire of most of the readers, formulas will be omitted as much as possible. Diagrams will illustrate the purposes of certain techniques. For a more technical, mathematical-oriented approach, reference is made to basic statistical textbooks (Cooley and Lohnes, 1971; Meisel, 1972). Of course, the criteria to define the groups to be investigated should be very clearly defined beforehand, irrespective of the nature of the criteria (morphological, biochemical, prognostic, or other).


Decision Tree Discriminant Analysis Decision Rule Linear Discriminant Analysis Receiver Operating Characteristic Curve 
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. Agrafojo Blanco, A., Gibbs, A.C.C. and Langley, F.A. (1977). Histological discrimination of malignancy in mucinous ovarian tumours. Histopathology 1, 431–443.PubMedCrossRefGoogle Scholar
  2. Baak, J.P.A. and Bezemer, P.D. (1974). Discriminant analysis of stereological parameters in normals and SIDS. In: Quantitative analysis of microstructures in medicine, H.E. Exner, editor, Riederer, Stuttgart, pp. 113–123.Google Scholar
  3. Baak, J.P.A., Diegenbach, P.C., Kurver, P.H.J., Stolk, J.G., Harten, J.J. van der (1979). An example of quantitative microscopy in individual patient care. Mikroskopie 37:305–307.Google Scholar
  4. Baak, J.P.A., Agrafojo Blanco, A., Kurver, P.H.J., Langley, F.A., Boon, M.E., Lindeman, J. and Diegenbach, P.C. (1981). Quantitation of borderline and malignant mucinous ovarian tumours. Histo-pathology 5, 353–360.Google Scholar
  5. Baak, J.P.A., Kurver, P.H.J., Snoo-Nieuwlaat, J.E. de, Graaf, S. de, Makkink, B., Boon, M.E. (1982). Prognostic indicators in breast cancer — morphometric methods. Histopathology 6, 327–339.PubMedCrossRefGoogle Scholar
  6. Bezemer, P.D., Baak, J.P.A., With, C. de (1977). Discriminant analysis exemplified with quantitative features of the endometrium. Eur. J. Obstet. Gynec. Reprod. Biol. 8:209–215.CrossRefGoogle Scholar
  7. Boon, M.E., Trott, P.A., Kaam H. van, Kurver, P.H.J., Leach, A. and Baak, J.P A. (1982 a). Morphometry and cytodiagnosis of breast lesions. Virchows Arch. (Pathol. Anat.) 369, 9–18.CrossRefGoogle Scholar
  8. Boon, M.E., Loewhagen, T., Lopes Cardozo, P., Blonk, D.I., Kurver, P.H.J, and Baak, J.P.A. (1982 b). Computation of pre-operative diagnosis probability for follicular adenoma and carcinoma of the thyroid on aspiration smears. Quant. Anal. Cytol. 4, 1–5.Google Scholar
  9. Carpenter, R.G. and Emery, J.L. (1974). The identification and follow-up of high risk infants. In: Sudden infant death syndrome, R.R. Robinson, editor, London and Toronto, Foundation for the study of infant death, pp. 91–96.Google Scholar
  10. Cooley, W.W. and Lohnes, P.R. (1971). Multivariate data analysis. John Wiley and Sons Inc., New York.Google Scholar
  11. Galen, R.S. and Gambino, S.R. (1975). Beyond normality: the predictive value and efficiency of medical diagnoses. John Wiley and Sons Inc., New York, pp. 10–14.Google Scholar
  12. Hart, W. and Norris, H.J. (1973). Borderline and malignant mucinous tumors of the ovary: histologic criteria and clinical behaviour. Cancer 31, 1031–1045.PubMedCrossRefGoogle Scholar
  13. Hermans J. and Habbema J.D.F. (1975). Comparison of five methods to estimate posterior probabilities. EDV in Medizin und Biologie 1, 14–19.Google Scholar
  14. Lusted, L.B. (1978). General problems in medical decision making with comment on ROC analyses. Semin. Nucl. Med. 8, 299–306.PubMedCrossRefGoogle Scholar
  15. Meisel, W.S. (1972). Computer-oriented approaches to pattern recognition. Mathematics in science and engineering, vol. 83, Academic Press, New York.Google Scholar
  16. Metz, C.E. (1978). Basic principles of ROC analyses. Semin. Nucl, Med., 8, 283–298.CrossRefGoogle Scholar
  17. Pauker, S.G. and Kassirer, J.P. (1980). The threshold approach to clinical decision making. N. Engl. J. Med. 302, 1109–1117.PubMedCrossRefGoogle Scholar
  18. Sappenfield, R.W., Beeler, M.F., Cartou, P.G. and Bordreau, D.A. (1981). Nine-cell diagnostic decision matrix. Am. J. Clin. Pathol. 75, 769–772.PubMedGoogle Scholar
  19. Schwartz, W.B., Wolfe, H.J. and Pauker, S.G. (1981). Pathology and probabilities. A new approach to interpreting and reporting biopsies. N. Engl. J. Med. 305, 917–923.PubMedCrossRefGoogle Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 1983

Authors and Affiliations

  • J. P. A. Baak
  • F. A. Langley
  • J. Hermans

There are no affiliations available

Personalised recommendations