Improving the Diagnostic Performance of MUNIN by Remodelling of the Diseases

  • Steen Andreassen
  • Marko Suojanen
  • Björn Falck
  • Kristian G. Olesen
Conference paper
Part of the Lecture Notes in Computer Science book series (LNCS, volume 2101)


The paper describes a revision of the structure of MUNIN, a causal probabilistic network that specifies a stochastic model of the relations between a range of neuromuscular diseases and the findings associated with these diseases. The stochastic model was revised 1) to achieve a more flexible specification of the anatomical distribution of findings associated with the diseases, 2) to allow diagnosis of diseases or groups of diseases (e.g. polyneuropathies and motor neuron diseases) previously lumped under the common concept of diffuse neuropathy and 3) to model the correlation between carpal tunnel syndrome on the left and right side. Minor adjustments were also made to some of the conditional probabilities used to specify the pathophysiology of the diseases.

The diagnostic capability of the revised model was evaluated by letting MUNIN diagnose 30 cases. The evaluation showed that the revised model performed better with a sensitivity of 94% and a specificity also of 94%.


Carpal Tunnel Syndrome Motor Neuron Disease Myotonic Dystrophy Nerve Lesion Motor Neuron Disease 
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.


Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.


  1. 1.
    Andreassen S, Jensen FV, Andersen SK, Falck B, Kjærulff U, Woldbye M, Sørensen AR, Rosenfalck A and Jensen F 1989. MUNIN — an expert EMG assistant. In: Computer-aided electromyography and expert systems, ed. JE Desmedt. Elsevier, Amsterdam, pp. 255–277.Google Scholar
  2. 2.
    Andreassen S, Jensen FV and Olesen KG 1991. Medical expert systems based on causal probabilistic networks. International Journal of Biomedical Computing 28: 1–30.CrossRefGoogle Scholar
  3. 3.
    Andreassen S, Falck B and Olesen KG 1992. Diagnostic function of the microhuman prototype of the expert system — MUNIN. Electroencephalography and Clinical Neurophysiology 85: 143–157.CrossRefGoogle Scholar
  4. 4.
    Andreassen S, Rosenfalck A, Falck B, Olesen KG and Andersen SK 1996. Evaluation of the diagnostic performance of the expert EMG assistant MUNIN. Electroencephalography and Clinical Neurophysiology 101: 129–144.CrossRefGoogle Scholar
  5. 5.
    Jensen FV 1996. An introduction to Bayesian networks. UCL Press.Google Scholar
  6. 6.
    Olesen KG, Kjærulff U, Jensen F, Jensen FV, Falck B, Andreassen S and Andersen SK 1989. A MUNIN network for the median nerve — a case study on loops. Applied Artificial Intelligence 3: 385–403.CrossRefGoogle Scholar
  7. 7.
    Olesen KG and Andreassen S 1996. A heuristic method for diagnosing multiple diseases in complex medical domains modelled by causal probabilistic networks. In: Medical Informatics Europe’ 96, eds. J Brender et al. IOS Press, pp. 614–618.Google Scholar
  8. 8.
    Pearl J 1988. Probabilistic reasoning in intelligent systems. Morgan Kaufmann.Google Scholar
  9. 9.
    Suojanen M, Olesen KG and Andreassen S 1997. A method for diagnosing in large medical expert systems based on causal probabilistic networks. In: Lecture Notes in Artificial Intelligence 1211, Proceedings of the 6th Conference on Artificial Intelligence in Medicine Europe (AIME’97), eds. E Keravnou et al. Springer, pp. 285–295.Google Scholar
  10. 10.
    Suojanen M, Andreassen S and Olesen KG 1999. The EMG diagnosis — an interpretation based on partial information. Medical Engineering & Physics 21: 517–523.CrossRefGoogle Scholar
  11. 11.
    Suojanen M, Andreassen S and Olesen KG 2001. A Method for Diagnosing Multiple Diseases in MUNIN. IEEE Transactions on Biomedical Engineering. In press.Google Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 2001

Authors and Affiliations

  • Steen Andreassen
    • 1
  • Marko Suojanen
    • 1
  • Björn Falck
    • 2
  • Kristian G. Olesen
    • 1
  1. 1.Department of Medical Informatics and Image Analysis, Institute of Electronic SystemsAalborg UniversityAalborgDenmark
  2. 2.Department of Clinical NeurophysiologyTurku University HospitalFinland

Personalised recommendations