Interpretable Fuzzy Modeling for Decision Support in IgA Nephropathy

摘要

The aim of the work is to show the potential usefulness of interpretable fuzzy modeling for decision support in medical applications. For this pursuit, we present an approach for designing interpretable fuzzy systems concerning the prognosis prediction in Immunoglobulin A Nephropathy (IgAN). To deal with such a hard problem, prognosis has been granulated into three classes; then, a number of fuzzy rule based classifiers have been designed so that several interpretability constraints are satisfied. The resulting classifiers have been evaluated in terms of classification accuracy (also compared with a standard neural network), some of interpretability indexes, and in terms of unclassified samples. Experimental results show that such models are capable to provide both a first estimation of prognosis and a readable knowledge base that can be inspected by physicians for further analyses.