Fuzzy Knowledge Adaptation

The goal is to define advanced techniques and approaches for tuning medical expert's knowledge, expressed in the form of linguistic variables, linguistic values and fuzzy rules, mainly when it is just partially or qualitatively formalized. Our approach, devised to classification problems in medicine, is planned for optimizing the shapes of the membership functions for each linguistic variable involved in the rules by applying an adaptive technique based on an evolutionary algorithm, i.e. Differential Evolution. The technique has been thought to guarantee the linguistic meaning and interpretability by meeting a set of constraints, such as distinguishability, orthogonality, normality and coverage, and to obtain conjunctly a good classification accuracy and a short execution time. In order to best fit the structure of medical knowledge and the peculiarities of the medical inference, we also proposed a fuzzy inference technique, resembling the Fuzzy Inference Ruled by Else-Action (FIRE) method, to face the lack of exclusionary rules in medical classification problems without requiring the medical practitioners to be upset in their way of thinking and working, that means without forcing them to write also rules for the negative evidence.


  • M. Esposito, I. De Falco, G. De Pietro, "An Evolutionary-Fuzzy Approach for building a DSS for Classification in Medical Problems", in 6th International Conference on Advanced Information Management and Service (IMS), Seoul, Korea November 30 - December 2, 2010, pp. 310 - 317.
  • M. Esposito, I. De Falco, G. De Pietro, "An Evolutionary-Fuzzy Approach for Supporting Diagnosis and Monitoring of Multiple Sclerosis" the 5th Cairo International Conference on Biomedical Engineering (CIBEC 2010), Cairo, Egypt, December 16-18, 2010, pp. 108 - 111.
  • I. De Falco, M. Esposito, G. De Pietro, "An advanced DSS for classification of multiple-sclerosis lesions in MR images", in CISSE10: Proceedings of the International Joint Conferences on Computer, Information, and Systems Sciences, and Engineering, Springer Netherlands, 2011.
  • M. Esposito, De Falco, G. De Pietro. An Evolutionary-Fuzzy DSS for Assessing Health Status in Multiple Sclerosis Disease. International Journal of Medical Informatics, vol. 80, no 12, pp. e245–e254, December 2011.