Acceleration and Deceleration Detection and Baseline Estimation

CTG monitoring of the fetal heart rate and uterine contractions during pregnancy and delivery provide information on the physiological condition of the fetus needed to identify hypoxia. Interpretation, however, requires experience and important patterns might be missed or misinterpreted due to stress, exhaustion or distraction. Interpretation has also been reported subjective with poor reproducibility. Computerized analysis of the fetal heart rate could help overcome these issues and provide fully reproducible analysis. The aim of this thesis is to evaluate algorithms for automated detection of fetal heart rate accelerations and decelerations and estimating the basal heart rate, for fetal monitoring during labor. Algorithms have been developed and the performance has been compared to the opinions of clinical experts and to other competing algorithms. Disagreement between clinical experts is considerably high for both acceleration and deceleration detection, but the observers are fairly consistent in estimating the basal heart rate. The developed algorithms all perform better than the competing algorithms and the performance is within or close to the uncertainty between the clinical experts.