Methods of Information in Medicine, Table of Contents

Methods Inf Med 1985; 24(01): 39-45
DOI: 10.1055/s-0038-1635355

Original Article

Schattauer GmbH

Selection and Assessment of Laboratory Tests for the Evaluation of Liver Functional Impairment

Auswahl und Beurteilung von Laboratoriumstests für die Bewertung von funktionellen Leberschäden

G. Belforte

¹(From the Dipartimento di Automatica e Informatica, Politecnico di Torino, II. Clinica Medica and Dipartimento di Informatica, Università di Torino, Italy)

,

B. Bona

¹(From the Dipartimento di Automatica e Informatica, Politecnico di Torino, II. Clinica Medica and Dipartimento di Informatica, Università di Torino, Italy)

,

C. Cravetto

¹(From the Dipartimento di Automatica e Informatica, Politecnico di Torino, II. Clinica Medica and Dipartimento di Informatica, Università di Torino, Italy)

,

S. Frediani

¹(From the Dipartimento di Automatica e Informatica, Politecnico di Torino, II. Clinica Medica and Dipartimento di Informatica, Università di Torino, Italy)

,

M. Milanese

¹(From the Dipartimento di Automatica e Informatica, Politecnico di Torino, II. Clinica Medica and Dipartimento di Informatica, Università di Torino, Italy)

,

M. Molino

¹(From the Dipartimento di Automatica e Informatica, Politecnico di Torino, II. Clinica Medica and Dipartimento di Informatica, Università di Torino, Italy)

,

L. Saitta

¹(From the Dipartimento di Automatica e Informatica, Politecnico di Torino, II. Clinica Medica and Dipartimento di Informatica, Università di Torino, Italy)

,

R. Tempo

¹(From the Dipartimento di Automatica e Informatica, Politecnico di Torino, II. Clinica Medica and Dipartimento di Informatica, Università di Torino, Italy)

› Author Affiliations

Summary

A large number of biochemical tests is at present available for the evaluation of the different aspects of the liver function. As a consequence, the requirement of laboratory investigations is generally increasing in clinical practice. However, the usefulness of such a tendency in liver investigation is questionable: the doubt concerns both the diagnostic efficacy of new individual tests and the opportunity of a progressive increase in tests requirement.

This paper deals with the problem of selecting liver tests on the basis of their information content with respect to different aspects of liver function. The study was performed on 288 normal and pathological subjects. Two different methods were used: a classical statistical feature selection method, and a method based on the probability theory developed in the framework of the fuzzy sets theory. The ranking of the tests was based on the estimated misclassification rate.

The results obtained appear quite relevant from a practical point of view: no effective improvement in liver functional assessment is reached by using more than 3 or 4 selected tests among the many available for each of the functional aspects considered.

A reasonable compromise between diagnostic efficacy, feasibility, and costs indicates that the following groups of tests could be proposed for the functional assessment of liver diseases in clinical routine: Albumin, Pseudo-cholinesterase, and Prothrombin Time for biosynthesis and metabolism; Bilirubin, Gamma-glutamyl-transpeptidase, Bile Acids (fasting) for cholestasis; Aspartate Aminotransferase, Alanine Aminotransferase, Isocitrate-dehydrogenase, for cytolysis; Gamma-globulin, G, A and M Immunoglobulin for reactivity.

Derzeit steht eine große Anzahl von biochemischen Tests für die Bewertung der verschiedenen’ Aspekte der Leberfunktion zur Verfügung. Dementsprechend nimmt die Nachfrage nach Laboratoriumsuntersuchungen in der klinischen Praxis allgemein zu. Die Nützlichkeit dieser Tendenz in Bezug auf die Untersuchung der Leber ist anzuzweifeln: dieser Zweifel betrifft sowohl die diagnostische Wirksamkeit neuer individueller Tests als auch die Zweckmäßigkeit einer ständigen Steigerung der Nachfrage nach solchen Tests.

Die vorliegende Arbeit behandelt das Problem der Auswahl von Lebertests auf der Grundlage ihres Informationsgehaltes in bezug auf verschiedene Aspekte der Leberfunktion. Die Studie wurde an 288 Normalpersonen und Kranken durchgeführt. Dabei wurden zwei verschiedene Methoden angewendet: eine klassische statistische Methode der Merkmalsauswahl und eine auf der Wahrscheinlichkeitstheorie basierende, im Rahmen der Fuzzy-sets-Theorie entwik-kelte Methode. Die Rangfolge der Tests basierte auf der geschätzten Fehlklassifikationsrate. Aus praktischer Sicht erscheinen die Ergebnisse recht bedeutsam: die Anwendung von mehr als 3 oder 4 ausgewählten Tests unter den vielen, die für jeden der betrachteten funktionellen Aspekte zur Verfügung stehen, ergab keine effektive Verbesserung in der Beurteilung der Leberfunktion.

Ein vernünftiger Kompromiß zwischen diagnostischer Effektivität, Machbarkeit und Kosten läßt folgende Gruppen von Tests für die funktionelle Beurteilung von Leberkrankheiten in der klinischen Praxis vorschlagen: Albumin, PseudoCholinesterase und Prothrombinzeit für Biosynthese und Stoffwechsel; Bilirubin, Gamma-Glutamyl-Transpeptidase und Gallensäuren (Fasten) für Cholestase; Aspartat-Aminotransferase, Alanin-Aminotransferase und Isoci-trate-Dehydrogenase für Cytolyse; Gamma-Globulin, G-, A- und M-Immunoglobuline für Reaktivität.

Key-Words:

Key-Words:

Liver Function Tests - Discriminant Analysis - Fuzzy Sets - Test Organization

Schlüssel-Wörter:

Schlüssel-Wörter:

Leberfunktionstests - Diskriminanzanalyse - Fuzzy Sets - Testorganisation

REFERENCES
1 Bezdek J, Castelaz P. Prototype Classification and Features Selection with Fuzzy Sets. IEEE Trans. Syst. Man Cybern SMC 1977; 7: 87-92.
2 De Mori R, Saitta L. Automatic Learning of Fuzzy Naming Relations over Finite Languages. Inform. Sci 1980; 21: 93-139.
3 Feinstein A. R. On. the Sensitivity, Specificity and Discrimination of Diagnostic Tests. St. Louis: C. V. Mosby; 1977
4 Fu K. S. Sequential Methods in Pattern Recognition and Machine Learning. New York: Academic Press; 1968
5 Fukunaga K. Introduction to Statistical Pattern Recognition. New York: Academic Press; 1972
6 Habbema J. D. F, Hermans J. Selection of Variables in Discriminant Analysis by F-Statistic and Error Rate. Technometrics 1977; 19: 487-493.
7 Hardison J. E. Compulsion to be Complete. New Engl. J. Med 1979; 300: 193-194.
8 Hermans J, Habbema J. D. E. Manual for the ALLOC Discriminant Analysis Programs. Techn. Report. Dept. of Medical Statistics, University of Leiden. 1976
9 Hills M. Allocation Rules and Error Rates. J. roy. Statist. Soc. B 1966; 28: 1-31.
10 Kassirer J. P, Pauker S. G. Should Diagnostic Testing be Regulated? New Engl. J. Med 1978; 299: 947-949.
11 Keeney R. L, Raiffa H. Decision with Multiple Objectives. Preferences and Values Tradeoffs. New York: J. Wiley; 1976
12 Lachenbruch P. A. On Expected Probabilities of Misclassification in Discriminant Analysis, Necessary Sample Size, and a Relation with the Multiple Correlation Coefficient. Biometrics 1968; 24: 823-834.
12a Lachenbruch P. A, Mickey M. R. Estimation of Error Rates in Discriminant Analysis. Technometrics 1968; 10: 1-11.
13 Lester E. The Machines are Taking over. The Lancet 1979; II: 949-950.
14 Martin S. P. The Clinical Laboratory: Cost Benefit and Effectiveness. Ann. intern. Med 1971; 75: 309-310.
15 Milanese M, Molino G. et al. Qualitative and Quantitative Evaluation of Liver Diseases. Proc. 4th Symp. on Computer Applications in Medical Care. Washington: 1980: 555-563.
16 Molino G, Cravetto C. Laboratory Tests Evaluation and Selection for Functional Assessment of Liver Diseases. Proc. 17th E.A.S.L. Meeting, Goteborg, Sweden, Sept. 1982
17 Parzen E. On Estimation of a Probability Density Function and Mode. Ann. math. Statist 1962; 33: 1065-1076.
18 Ransohoff D. F, Feinstein A. R. Problems of Spectrum and Bias in Evaluating the Efficiency of Diagnostic Tests. New Engl. J. Med 1978; 299: 926-929.
19 Russe H. The Use and Abuse of Laboratory Tests. Med. Clin. N. Amer 1969; 53: 225-231.
20 Saitta L, Gaia E. Fuzzy Cluster Set for Medical Diagnosis. Proc. IEEE Int. Conf. on Cybernetics and Society. Tokyo: 1978: 248-253.
21 Wagner G, Tautu P, Wolber U. Problems of Medical Diagnosis – A Bibliography. Meth. Inform. Med 1978; 17: 55-74.
22 A Survey of Opportunities and Needs in Research on Digestive Diseases. Work Group V: Metabolic Diseases of the Liver. Gastroenterology 1975; 69: 1108-1120.
23 Issue on Decision Making in Health Care. New Engl. J. Med 1975; 293: 211-244.
24 Special Issue on Statistical Methods for Clinical Decision Making. J. roy. Coll. Phys. Lond 1975; 9: 191-269.