Extracting fetal electrocardiogram signal using ANFIS trained by genetic algorithm

This paper uses a method for extracting the Fetal Electrocardiogram (FECG) signal from two ECG signals recorded at thoracic and abdominal areas of mother. The thoracic ECG is assumed to be completely maternal ECG (MECG) while the abdominal ECG is assumed to be a combination of mother's and fetus's ECG signals and random noise. The maternal component of the abdominal ECG is a nonlinearly transformed version of MECG. The method uses Adaptive Neuro-Fuzzy Inference System (ANFIS) structure to identify the nonlinear transformation. We have used Genetic Algorithm (GA) as a tool for training the ANFIS structure. By identifying the nonlinear transformation, we have extracted FECG by subtracting the aligned version of the MECG signal from the abdominal ECG (AECG) signal. We validate the method on both real and synthetic ECG signals. The results show improvement in extraction of FECG signal with the method in this study.