A case study on medical diagnosis of cardiovascular diseases using a Genetic Algorithm for Tuning Fuzzy Rule-Based Classification Systems with Interval-Valued Fuzzy Sets

In this contribution, we use Fuzzy Rule-Based Classification Systems for classifying the patients with respect to the risk of suffering cardiovascular diseases. Specifically, we use a methodology in which the linguistic labels of the classifier are modeled by means of IVFSs. Thereafter, they are genetically post-processed for tuning the amplitude of the support of the upper bound of each membership function. In this manner a good management of the uncertainty, associated with the definition of the fuzzy terms, is provided to the system. We show the goodness of our methodology by comparing its performance with respect to the one provided by the initial system in this specific medical case. First, we study the global classification improvement and then, we carry out an exhaustive analysis of the behavior of our approach in which we observe the enhancement achieved in several specific situations.