gms | German Medical Science

62. Jahrestagung der Deutschen Gesellschaft für Medizinische Informatik, Biometrie und Epidemiologie e. V. (GMDS)

Deutsche Gesellschaft für Medizinische Informatik, Biometrie und Epidemiologie

17.09. - 21.09.2017, Oldenburg

Guided Expert Modeling of Clinical Bayesian Network Decision Graphs

Meeting Abstract

  • Mario A. Cypko - Innovation Center Computer Assisted Surgery, University of Leipzig, Faculty of Medicine, Leipzig, Germany, Leipzig, Deutschland
  • Matthäus Stöhr - Universität Leipzig Medizinische Fakultät, Leipzig, Deutschland
  • Steffen Oeltze-Jafra - Innovation Center Computer Assisted Surgery (ICCAS), Universität Leipzig, Leipzig, Deutschland
  • Andreas Dietz
  • Heinz U. Lemke

Deutsche Gesellschaft für Medizinische Informatik, Biometrie und Epidemiologie. 62. Jahrestagung der Deutschen Gesellschaft für Medizinische Informatik, Biometrie und Epidemiologie e.V. (GMDS). Oldenburg, 17.-21.09.2017. Düsseldorf: German Medical Science GMS Publishing House; 2017. DocAbstr. 127

doi: 10.3205/17gmds190, urn:nbn:de:0183-17gmds1905

Veröffentlicht: 29. August 2017

© 2017 Cypko et al.
Dieser Artikel ist ein Open-Access-Artikel und steht unter den Lizenzbedingungen der Creative Commons Attribution 4.0 License (Namensnennung). Lizenz-Angaben siehe



Increasing knowledge leads to more complex decision making. In previous work, we developed a system that supports clinical experts in the decision-making process. The system is built upon probabilistic networks, which are particularly suited for modeling complex decisions, and can comprise all relevant characteristics of a disease, examination methods, therapy options, and therapy effects. Based on the model and a set of observations (patient data), an inference algorithm computes the likelihood of occurrence for the model’s unobserved characteristics (e.g., missing examination, treatment decisions, and potential outcomes). However, modeling a complex decision requires teamwork of domain experts and knowledge engineers making the process expensive, time-consuming, and prone to misunderstandings. We present a novel guided method that enables domain experts for autonomous modeling. We demonstrate our approach by an example of middle ear infection.

Die Autoren geben an, dass kein Interessenkonflikt besteht.

Die Autoren geben an, dass kein Ethikvotum erforderlich ist.


Kohn KT, Corrigan JM, Donaldson MS. To Err Is Human Building: a Safer Health System. Washington, DC: National Academy Press; 1999.
Seixas FL, Zadrozny B, Laks J, Conci A, Saade DCM. A Bayesian network decision model for supporting the diagnosis of dementia, Alzheimer׳ s disease and mild cognitive impairment. Computers in biology and medicine. 2014;51:140-158.
Pearl J. Causality. Cambridge University Press; 2009.
Druzdzel MJ, Onisko A, Schwartz D, Dowling JN, Wasyluk H. Knowledge engineering for very large decision-analytic medical models. AMIA; 1999.
Onisko A. Probabilistic causal models in medicine: Application to diagnosis of liver disorders [Ph.D. dissertation]. Warsaw: Institute of Biocybernetics and Biomedical Engineering, Polish Academy of Science; 2003.
Lemke HU. Interactive graphics in an intergraded computer aided design system [Ph.D. dissertation]. Cambridge: Computer Laboratory, University of Cambridge; 1972.
Cypko MA, Stoehr M, Denecke K. Web-based guiding of clinical experts through the modeling of treatment decision models using MEBN with an example of laryngeal cancer. Int J CARS. 2015;10(1).
Cypko MA, Hirsch D, Koch L, Stoehr M, Strauss G, Denecke K. Web-tool to support medical experts in probabilistic modelling using large Bayesian networks with an example of rhinosinusitis. Stud Health Technol Inform. 2014;216:259-263.
Cypko MA, Stoehr M, Kozniewski M, Druzdzel MJ, Dietz A, Lemke HU. Validation workflow for a clinical Bayesian network model in multidisciplinary decision-making in head and neck oncology treatment. Int J Comput Assist Radiol Surg. 2017.