The goal of the project is development of a predictive model
and identification of biomarkers to assess the risk of preterm
birth. Using signal processing methods to analyze uterine
records (electrohysterogram, EHG, signals accompanied by a
simultaneously recorded external tocogram, TOCO, measuring
mechanical uterine activity), and using machine learning methods,
we estimate the degree of the risk of spontaneous preterm birth.
Researches include development of software for visualizing the
signals and expert annotating of the records, characterization
and estimation of the signals using linear and non-linear signal
processing techniques in time and frequency domain, mathematically
modeling the electro-mechanical activities of the uterus, and
development of automatic methods for predicting preterm birth.