Research program P3-0124 (2014 - 2020) is funded by the ARRS, Slovenian research agency.
Research partner: University Medical Center, Ljubljana, Slovenia.
The goal of the project is to develop a predictive model to assess the risk of preterm delivery. Using signal analysis of uterine records (electrohysterogram, EHG, signals accompanied by a simultaneously recorded external tocogram, TOCO, measuring mechanical uterine activity), we estimate the risk for preterm delivery. We are developing a new database that will contain uterine records recorded twice
given pregnancy (around 23rd and around 31st week of pregnancy), with spontaneous (preterm and term), induced, and cesarean
section deliveries. We characterize the records, mathematically model the electro-mechanical activity of uterus, and develop automatic methods for predicting preterm birth. Signal analysis include linear time- and frequency-domain techniques, signal power spectra estimation, and non-linear techniques such as sample entropy.