Automatic Analysis of Intrapartum Fetal Heart Rate

Doctoral thesis statement for obtaining the academic title of " Doctor " , Resume 23 Shrnutí 24 1 1 Problem introduction Fetal heart activity is the prominent source of information about the fetal well being during delivery. Cardiotocography (CTG) – recording of fetal heart rate (fHR) and uterine contractions (TOCO) enables obstetricians to detect possible ongoing fetal hypoxia, which may occur during delivery even in a previously uncomplicated pregnancy. Figure 1 shows an example of the CTG recording. Even though fetus has its own natural defense mechanism to tackle the oxygen insufficiency during delivery, in some cases, only timely intervention can prevent adverse consequences [1]. Hypoxia, with prevalence lying in the region of 0.6% [21] to 3.5% [34], is considered to be the third most common cause of newborn death [9] in developed countries. Instrumental evaluation of the fetal well-being during delivery is more than hundred years old. Auscultation – sensing of the fetal heart rate (fHR) using a fetal stethoscope – introduced by Pinard in 1876 – was replaced in the decades after the second world war by electronic fetal monitoring (EFM) with cardiotocography as the most important representant. Cardiotocography was introduced in late 1960s and is still the most prevalent method of intrapartum hypoxia detection. It did not, however, bring the expected improvements in the delivery outcomes in comparison to previously used intermittent auscultation [33] and, moreover, continuous CTG is one of the main suspects for increased rate of cesarean sections for objective reasons [33]. To improve the results of cardiotocography, International Federation of Gynecology and Obstetrics (FIGO) introduced general guidelines [13] for CTG assessment. They are based on evaluation of macroscopic morphological fHR features and their relation to the tocographic measurement. Even though the guidelines are available for more than twenty years poor interpretation of CTG still persists [3] with large inter-observer as well as intra-observer assessment variations [6, 4]. Therefore one of this thesis' goals is to contribute to the discussion about the feasibility of the automatic evaluation of the fHR and to the possibility of improving the inter-observer and intra-observer variability. Recently ST-analysis has been getting much attention as an extension of the classical CTG measurements using additional information from the invasive measurement of the fetal ECG [28]. Although most studies show that ST-analysis is performing slightly better [2], it is important to keep in mind, that the first step to correctly interpret the ST …