Early diagnosis of threatened premature labor by electrohysterographic recordings – The use of digital signal processing

Abstract Prevention and early diagnosis of imminent preterm labor are considered to be the most important perinatal challenge nowadays. Significant progress has been observed on postnatal care of premature infants, but without reducing the prevalence of preterm delivery. Our study was focused on comparison of three methods of spectral analysis of electrohysterographic (EHG) signals: fast Fourier transform (FFT), wavelet transform (WT) and autoregressive modeling (AR). Complexity of the electrohysterographic signals was analyzed by using: the approximate entropy (ApEn), Lempel–Ziv complexity measure (L–Z). Additionally, the work evaluated the applicability of EHG in diagnosing imminent premature labor. EHG signals were recorded among 60 patients with threatened preterm labor symptoms between the 24th and 34th week of pregnancy. Patients included to the study had a shortened cervix (less than 20 mm) without regular uterine contractions recorded on regular cardiotocography (CTG). The women were divided into two groups: those delivering within 7 days – group A ( n  = 15) and women delivering after 7 days – group B ( n  = 45). The study confirmed differences in bioelectrical activity of uterus between patients delivering prematurely within 7 days and after from the EHG registration for all analyzed methods.