Expert System Design in Hematology Diagnosis

Abstract:

A two-part study was designed to test the hypothesis that sufficient information is available from a modern hematology analyzer (the Coulter STKS) to reach a reliable intermediate conclusion which can be used as input to the next decision-making level in the design of a high-performance expert system for hematology diagnosis. In phase one, we analyzed the performance of three probabilistic systems (using Bayes’ rule) which interpret STKS data: a control system which took the traditional approach of classifying cases into specific diagnoses, and two test systems which were designed to reach only an intermediate conclusion but not a final diagnosis. One of the test systems classified cases into “textbook categories” of disease and the other utilized defined diagnostic patterns. The systems were tested with 150 cases. The pattern approach ranked the correct choice first in 141 of 150 cases (94%). In phase two, we abandoned Bayes’ rule, reformulated the pattern approach into a heuristic classification system, and tested its reliability on 820 cases. The algorithm of the reformulated system was able to classify all 820 cases into the same predominant pattern as a panel of three experienced laboratory hematologists.

Keywords

Expert System - Hematology - Pattern Analysis - Bayesian Probability - Heuristic Classification

• REFERENCES

• 1 Sultan C, Imbert M, Priolet G. Decisionmaking system (DMS) applied to hematology. Diagnosis of 180 cases of anemia secondary to a variety of hematologic disorders. Hematol Pathol 1988; 02: 221-8.
• 2 Blomberg DJ, Ladley JL, Fattu JM, Patrick EA. The use of an expert system in the clinical laboratory as an aid in the diagnosis of anemia. Am J Clin Pathol 1987; 87: 608-13.
• 3 Bates JE, Bessman JD. Evaluation of BCDE, a microcomputer program to analyze automated blood counts and differentials. Am J Clin Pathol 1987; 88: 314-23.
• 4 Quaglini S, Stefanelli M, Barosi G, Berzuini A. A performance evaluation of the expert system ANEMIA. Comput Biomed Res 1988; 21: 307-23.
• 5 Lanzola G, Stefanelli M, Barosi G, Magnani L. NEOANEMIA: A knowledge-based system emulating diagnostic reasoning. Comput Biomed Res 1990; 23: 560-82.
• 6 Tolmie CJ, du Plessis JP, Badenhorst PN. An expert system for the interpretation of full blood counts and blood smears in a hematology laboratory. Artif Intel Med 1991; 03: 271-85.
• 7 Alvey PL, Myers CD, Greaves MF. High performance for expert systems: I. Escaping from the demonstrator class. Med Inform 1987; 12: 85-95.
• 8 Warner BA, Reardon DM. A field evaluation of the Coulter STKS. Am J Clin Pathol 1991; 95: 207-17.
• 9 Beck JR. Interpretative reporting using decision science techniques. Lab Med 1991; 22: 712-7.
• 10 Sigaux F, Imbert M, Priolet G, Bucquen JJ, Levy C, Sultan C. Aide à la dècision en hèmatologic Caractèristiques et performances du programme. Deux cents cas d’anèmie. Presse Mèd 1987; 16: 111-4.
• 11 Milton JS, Tsokos JO. Statistical Methods in the Biological and Health Sciences. New York: McGraw-Hill; 1983: 227-31.
• 12 Fryback DG. Bayes’ theorem and conditional nonindependence of data in medical diagnosis. Comput Biomed Res 1978; 11: 423-34.
• 13 Szolovits P, Pauker SG. Categorical and probabilistic reasoning in medical diagnosis. Artif Intel 1978; 11: 115-44.
• 14 Moskowitz AJ, Kuipers BJ, Kassirer JP. Dealing with uncertainty, risks, and tradeoffs in clinical decisions: A cognitive science approach. Ann Intern Med 1988; 108: 435-49.
• 15 Bartels PH. The diagnostic pattern in histopathology. Am J Clin Pathol 1989; 91 (Suppll): S7-S13.
• 16 Diamond LW, Nguyen DT, Priolet G, Sultan C. An expert system for the analysis of the hemogram and peripheral blood smear: A pattern approach. Blood 1991; 78 (Suppl. 01) 101A.