Evaluating Tests in Medical Diagnosis: Combining Machine Learning with Game-Theoretical Concepts

  • Karlson PfannschmidtEmail author
  • Eyke Hüllermeier
  • Susanne Held
  • Reto Neiger
Conference paper
Part of the Communications in Computer and Information Science book series (CCIS, volume 610)


In medical diagnosis, information about the health state of a patient can often be obtained through different tests, which may perhaps be combined into an overall decision rule. Practically, this leads to several important questions. For example, which test or which subset of tests should be selected, taking into account the effectiveness of individual tests, synergies and redundancies between them, as well as their cost. How to produce an optimal decision rule on the basis of the data given, which typically consists of test results for patients with or without confirmed health condition. To address questions of this kind, we develop an approach that combines (semi-supervised) machine learning methodology with concepts from (cooperative) game theory. Roughly speaking, while the former is responsible for optimally combining single tests into decision rules, the latter is used to judge the influence and importance of individual tests as well as the interaction between them. Our approach is motivated and illustrated by a concrete case study in veterinary medicine, namely the diagnosis of a disease in cats called feline infectious peritonitis.


Generalization Performance Cooperative Game Theory Combine Test Interaction Index Isotonic Regression 
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.


  1. 1.
    Addie, D.D., Paltrinieri, S., Pedersen, N.C.: Recommendations from workshops of the second international feline coronavirus/feline infectious peritonitis symposium. J. Feline Med. Surg. 6(2), 125–130 (2004)CrossRefGoogle Scholar
  2. 2.
    Benetka, V., Kübber-Heiss, A., Kolodziejek, J., Nowotny, N., Hofmann-Parisot, M., Möstl, K.: Prevalence of feline coronavirus types I and II in cats with histopathologically verified feline infectious peritonitis. Vet. Microbiol. 99(1), 31–42 (2004)CrossRefGoogle Scholar
  3. 3.
    Block, H., Qian, S., Sampson, A.: Structure algorithms for partially ordered isotonic regression. J. Comput. Graph. Stat. 3(3), 285–300 (1994)MathSciNetGoogle Scholar
  4. 4.
    Giori, L., Giordano, A., Giudice, C., Grieco, V., Paltrinieri, S.: Performances of different diagnostic tests for feline infectious peritonitis in challenging clinical cases. J. Small Anim. Pract. 52(3), 152–157 (2011)CrossRefGoogle Scholar
  5. 5.
    Grabisch, M., Nguyen, H., Walker, E.: Fundamentals of Uncertainty Calculi with Applications to Fuzzy Inference. Kluwer Academic Publishers, Dordrecht (1995)CrossRefzbMATHGoogle Scholar
  6. 6.
    Hartmann, K., Binder, C., Hirschberger, J., Cole, D., Reinacher, M., Schroo, S., Frost, J., Egberink, H., Lutz, H., Hermanns, W.: Comparison of different tests to diagnose feline infectious peritonitis. J. Vet. Intern. Med. 17(6), 781–790 (2003)CrossRefGoogle Scholar
  7. 7.
    Hirschberger, J., Hartmann, K., Wilhelm, N., Frost, J., Kraft, W.: Using direct immunofluorescence to detect coronaviruses in peritoneal in peritoneal and pleural effusions. Tierärztliche Praxis 23, 92–99 (1995)Google Scholar
  8. 8.
    Hirschberger, J., DeNicola, D.B., Hermanns, W., Kraft, W.: Sensitivity and specificity of cytologic evaluation in the diagnosis of neoplasia in body fluids from dogs and cats. Vet. Clin. Pathol. 28(4), 142–146 (1999)CrossRefGoogle Scholar
  9. 9.
    Jeffery, U., Deitz, K., Hostetter, S.: Positive predictive value of albumin: globulin ratio for feline infectious peritonitis in a mid-western referral hospital population. J. Feline Med. Surg. 14(12), 903–905 (2012)CrossRefGoogle Scholar
  10. 10.
    Kipar, A., Köhler, K., Leukert, W., Reinacher, M.: A comparison of lymphatic tissues from cats with spontaneous feline infectious peritonitis (FIP), cats with FIP virus infection but no FIP, and cats with no infection. J. Comp. Pathol. 125(2), 182–191 (2001)CrossRefGoogle Scholar
  11. 11.
    Lipovetsky, S., Conklin, M.: Analysis of regression in game theory approach. Appl. Stochast. Models Bus. Ind. 17(4), 319–330 (2001)MathSciNetCrossRefzbMATHGoogle Scholar
  12. 12.
    Loog, M.: Contrastive pessimistic likelihood estimation for semi-supervised classification. IEEE Trans. Pattern Anal. Mach. Intell. 38(3), 462–475 (2016)CrossRefGoogle Scholar
  13. 13.
    Murofushi, T., Soneda, S.: Techniques for reading fuzzy measures (iii): interaction index. In: 9th Fuzzy System Symposium, Sapporo, Japan, pp. 693–696 (1993)Google Scholar
  14. 14.
    Pardalos, P., Xue, G.: Algorithms for a class of isotonic regression problems. Algorithmica 23(3), 211–222 (1999)MathSciNetCrossRefzbMATHGoogle Scholar
  15. 15.
    Parodi, M.C., Cammarata, G., Paltrinieri, S., Lavazza, A., Ape, F.: Using direct immunofluorescence to detect coronaviruses in peritoneal and pleural effusions. J. Small Anim. Pract. 34(12), 609–613 (1993)CrossRefGoogle Scholar
  16. 16.
    Ritz, S., Egberink, H., Hartmann, K.: Effect of feline interferon-omega on the survival time and quality of life of cats with feline infectious peritonitis. J. Vet. Intern. Med. 21(6), 1193–1197 (2007)CrossRefGoogle Scholar
  17. 17.
    Shapley, L.: A value for n-person games. Ann. Math. Stud. 28, 307–317 (1953)MathSciNetzbMATHGoogle Scholar
  18. 18.
    Soma, T., Wada, M., Taharaguchi, S., Tajima, T.: Detection of ascitic feline coronavirus RNA from cats with clinically suspected feline infectious peritonitis. J. Vet. Med. Sci. 75(10), 1389–1392 (2013)CrossRefGoogle Scholar

Copyright information

© Springer International Publishing Switzerland 2016

Authors and Affiliations

  • Karlson Pfannschmidt
    • 1
    Email author
  • Eyke Hüllermeier
    • 1
  • Susanne Held
    • 2
  • Reto Neiger
    • 2
  1. 1.Department of Computer SciencePaderborn UniversityPaderbornGermany
  2. 2.Small Animal ClinicJustus-Liebig University GießenGiessenGermany

Personalised recommendations