Detection of fetal ECG with IIR adaptive filtering and genetic algorithms

The continuous monitoring of the fetal heart condition during pregnancy and labor is of great clinical importance. The cardiac electrical activity of the fetus (FECG) may be recorded by means of surface abdominal electrodes. The signal is severely contaminated by the maternal cardiac signal (MECG). FECG enhancement is usually performed by FIR adaptive filtering. A new IIR FECG enhancement system is suggested and evaluated. In order to avoid convergence into local extrema, the system employs a genetic algorithm (GA). Two architectures are considered. The first is a combination of an adaptive filter and GA where the GA is recruited whenever the adaptive filter is suspected of reaching a local extremum. The second is an independent GA search. The hybrid IIR-GA was shown to be superior to the conventional FIR adaptive filtering