Spatio-temporal signal processing for fetus and mother state monitoring during pregnancy

The premature birth is an extremely important problem of pregnancy. It is a major cause of long-term baby's morbidity and neonatal deaths. The currently used instrumental methods and devices for predicting preterm delivery are uncomfortable, inaccurate, and may be used only in clinic. Analysis of the electrohysterographic (EHG) signals, observed on abdominal surface of pregnant women, is a promising diagnostic tool for predicting preterm delivery. The most sensitive parameter of EHG signal, in this sense, is conduction velocity (CV) of EHG bursts that occur during the uterus contraction. In this paper was synthesized algorithm for CV estimating on the base of spatio-temporal processing of multichannel abdominal signal, and were performed CV measurements using real EHG recordings.