Combining ε-similar Fuzzy Rules for Efficient Classification of Cardiotocographic Signals

CardioTocoGraphic (CTG) monitoring is the primary method of fetal condition assessment. Due to the inter- and intra-observer disagreement between experts when evaluating signals visually, a well established solution supporting the diagnostic decision is automated classification of CTG signals. The goal of this paper is to propose a method of simplifying the fuzzy classifier rule base by combining ε-similar rules, to achieve high quality of CTG signals classification, but with fewer conditional rules. The results of experiments performed using the benchmark CTG database confirm the efficiency of the introduced method.