Evolving fuzzy neural network based on null-unineurons for the identification of coronary artery disease

Coronary diseases affect a large part of the world population and have become the target of significant research in the academic field. The creation and use of intelligent models to facilitate the diagnosis of these diseases can allow treatments to be performed promptly to avoid further problems for patients. This paper applies an innovative evolving fuzzy neural network model to solve the problem of coronary heart disease diagnosis and extract valuable insights from the evaluated dataset. The null-unineurons that compose the model's architecture can extract fuzzy rules, representing linguistic knowledge about the target problem. A dataset that condenses the most famous data sources on this problem to classify coronary heart disease was applied to state-of-the-art models of evolving fuzzy systems. The results obtained by the model applied in this study are similar to the state-of-the-art results. Furthermore, the model provides relevant interpretations about the evolution of the problem evaluation.