Diagnoses Generated by Numerical Taxonomic Methods Applied to Standard Blood Variables

Data taken from the blood of 262 patients diagnosed for malabsorption, elective cholecystectomy, acute cholecystitis, infectious hepatitis, liver cirrhosis, or chronic renal disease were analyzed with three numerical taxonomy (NT) methods : cluster analysis, principal components analysis, and discriminant function analysis. Principal components analysis revealed discrete clusters of patients suffering from chronic renal disease, liver cirrhosis, and infectious hepatitis, which could be displayed by NT clustering as well as by plotting, but other disease groups were poorly defined. Sharper resolution of the same disease groups was attained by discriminant function analysis.

Daten aus dem Blut von 262 Patienten mit den Diagnosen mangelhafte Absorption, Cholezystektomie, akute Cholezystitis, infektiöse Hepatitis, Leberzirrhose oder chronische Nierenerkrankung wurden mit Hilfe von drei numerischen Taxonomie(NT)-Methoden analysiert: Cluster-Analyse, Hauptkomponenten-Analyse und Diskrirninanz-Funktionsanalyse. Die Analyse der Hauptkomponenten ergab diskrete Cluster von Patienten mit chronischen Nierenleiden, Leberzirrhose und infektiöser Hepatitis, welche sowohl durch NT-Clustering als auch durch Plotting dargestellt werden konnten, während andere Patientengruppen schlecht definiert waren. Eine schärfere Abgrenzung der gleichen Patientengruppen wurde durch die Diskrirninanz-Funktionsanalyse erreicht.

Key-Words:

Computer-assisted Diagnosis - Cluster Analysis - Discriminant Function Analysis - Principal Component Analysis - Standard Blood Variables

Schlüssel-Wörter:

Computerunterstützte Diagnose - Clusteranalyse - Diskrirninanz-Funktionsanalyse - Hauptkomponenten-Analyse - standardisierte Blutvariable

• References

• 1 Alpébovitch A, Leminor M, Lellouch J. The determination of sequential strategies of diagnostic tests : Comparison of different methods by simulation. Meth. Inform. Med 1979; 18: 75-79.
• 2 Alpérovitch A, Fragu P. A suggestion for an effective use of a computer-aided diagnosis system m screening for hyperthyroidism. Meth. Inform. Med 1977; 16: 93-95.
• 3 Baron D. N, Fraser P. M. Medical applications of taxo-nomic methods. Brit. med. Bull 1968; 24: 236-240.
• 4 Buttons K. J, Gambino S. R. Laboratory diagnosis by computer. Comput. Biol. Med 1973; 3: 131-136.
• 5 Clifford H. T, Stephenson W. An Introduction to Numerical Classification. New York: Academic Press; 1975
• 6 Croft D. J. Is computerized diagnosis possible ? Comput. biomed. Res 1972; 5: 351-367.
• 7 Duran B. S, Lewis T. O. An application of cluster analysis to the construction of a diagnostic classification. Comput. Biol. Med 1974; 4: 183-188.
• 8 Ellis G, Goldberg D. M. Discriminant function analysis applied to laboratory tests in patients with hepatobiliary disease. Comput. biomed. Res 1979; 12: 483-501.
• 9 Fraser P. M. Precision in Diagnosis : Numerical Taxonomy and Discriminant Analysis. In Atkins H. J. B. (Edit.) : Measurement and Precision in Surgery Oxford: Blackwell; 1969
• 10 Friedman H. P, Goldwyn R. M, Siegel J. H. The use and interpretation of multivariate methods in the classification of stages in serious infectious disease processes in the critically ill. Perspeet. Biol. Med 1975; 1: 81-122.
• 11 Geer Van de J. P. Introduction to Multivariate Analysis for the Social Sciences. San Francisco: W. H. Freeman and Co; 1971
• 12 Goldberg D. M, Ellis G. Mathematical and computer-assisted procedures in the diagnosis of liver and biliary tract disorders. Advonc. clin. Chem 1978; 20: 49-128.
• 13 Gorry G. A, Barnett G. O. Experience with a model of sequential diagnosis. Comput. biomed. Res 1968; 1: 490-507.
• 14 Grams R. R, Johnson E. A, Benson E. S. Laboratory data analysis: Section III—Multivariate normality. Amer. J. clin. Path 1972; 58: 188-200.
• 15 Grams R. R, Johnson E. A, Benson E. S. Laboratory data analysis system: Section IV—Multivariate diagnosis. Amer. J. clin. Path 1972; 58: 201-215.
• 16 Hay R. W, Whitehead R. G. Serum-albumin as a prognostic indicator in oedematous malnutrition. Lancet 1975; II: 427-429.
• 17 Hockstra D. J. Sequential games and medical diagnosis. Comput. biomed. Res 1976; 9: 205-215.
• 18 Hughes W. L, Kalbfleisch J. M, Brandt Jr. E. N, Costiloe J. P. Myocardial infarction prognosis by discriminant analysis. Arch. int. Med 111 1963; 338-345.
• 19 Jardine N, Sibson R. Mathematical Taxonomy. New York—London: Wiley; 1971
• 20 Knapp R. G, Levi S, Lurie D, Westphal M. A computer-generated diagnostic guide: a comparison of statistical diagnosis and clinical diagnosis. Comput. Biol. Med 1977; 7: 223-230.
• 21 Ledley R. S, Lusted L. B. Reasoning foundations of medical diagnosis. Science 1959; 130: 9-21.
• 22 Mcneil B. J, Sherman H. Example: Bayesian calculations for the determination of the etiology of pleuritic chest pain in young adults in a teaching hospital, Part B. Comput. biomed. Res 1978; 11: 187-194.
• 23 Morrison D. R. Multivariate Statistical Methods. New York: Mcgraw-Hill; 1967
• 24 Reece R. L, Hobbie R. K. Computer evaluation of chemistry values: a reporting and diagnostic aid. Amer. J. clin. Path 1971; 57: 664-675.
• 25 Shea G. An analysis of the Bayes’ procedure for diagnosing multistage diseases. Comput. biomed. Res 1978; 11: 65-75.
• 26 Sneath P.H.A. The application of numerical taxonomy to medical problems. In Mathematics and Computer Science in Biology and Medicine London: H. M. S. O; 1965: 81-91.
• 27 Sneath P. H. A, Sokal R. R. Numerical Taxonomy. San Francisco: Freeman and Co; 1973
• 28 Sokal R. R, Sneath P.H.A. Principles of Numerical Taxonomy. San Francisco: Freeman and Co; 1962
• 29 Sokal R. R. Numerical Taxonomy and Disease Classification. In Jacquez J. A. (Edit.): The Diagnostic Process Ann Arbor: Malloy Lithographing; 1964: 51-68.
• 30 Spicer C. C, Jones J. H, Jones J. E.L. Discriminant and Bayes analysis in the differential diagnosis of Crohns disease and proctocolitis. Meth. Inform. Med 1973; 12: 118-122.
• 31 Warner H. R, Tobonto A. F, Veasey L. G. A mathematical approach to medical diagnosis. J. Amer. med. Ass 1961; 177: 177-183.
• 32 Wartak J. Mathematical model for medical diagnosis. Comput. Biol. Med 1974; 4: 79-84.
• 33 Winkel P. Patterns and clusters — Multivariate approach for interpreting clinical chemistry results. Clin. Chem 1973; 19: 1329-1338.
• 34 Winkel P, Paldam M, Tygstrup N. Numerical taxonomic analysis of symptoms and signs in 400 patients with cirrhosis of the liver. Comput. biomed. Res 1971; 3: 657-665.
• 35 Winkel P, Tygsteup N. Numerical taxonomic analysis of cirrhosis. II. Clinical significance of classifications. Comput. biomed. Res 1974; 7: 117-126.
• 36 Winkel P. (and Copenhagen Study Group for Liver Diseases): Numerical taxonomic analysis of cirrhosis. I. The Effect of the number and type of variables used. Comput. biomed. Res 1974; 7: 100-110.
• 37 Zajicek G, Maayan Ch., Rosenmann E. The application of cluster analysis to glomerular histopathy. Comput. biomed. Res 1977; 10: 471-481.
• 38 Zoltie N, Horrocks J. C, de Dombal F. T. Computer-assisted diagnosis of dyspepsia. Meth. Inform. Med 1977; 16: 89-92.