Cardiotocographic signals classification based on clustering and fuzzy if-then rules

Cardiotocographic (CTG) monitoring relies on analysis of fetal heart rate, fetal movements and uterine contractions. It is the most popular method for assessment of the fetal state. However, the diagnosis based on visual assessment of signals is very difficult and subjective. The aim of proposed work was to classify CTG signals for predicting the fetal outcome with a help of fuzzy if-then rules. The rules were created basing on a new fuzzy clustering method. The most of results obtained using 100-fold cross-validation procedure were better in comparison to the Lagrangian SVM method, which is one of best classifiers. The achieved classification error at the level below 21% and sensitivity equal to 77% seem to be encouraging.