Automatic analysis of the fetal heart rate variability and uterine contractions

The aim of this study is the identification of the dynamic relationship between intrapartum fetal heart rate (FHR) variability and uterine pressure (UP) for normal and hypoxic fetuses. A new method for automatic analysis and identification of uterine contractions is developed and tested on 552 recordings. Using contractions, the fetal heart rate variations are established: accelerations and decelerations (early, late and prolonged). These allow a classification of the recordings analyzed into two categories: normal and pathological. The results are compared to documented fetal outcome results quantified by means of low pH and Apgar values. The developed method provides a series of quantitative measurements for the uterine contractions signal (total number of contractions, total contraction time, total percentage of contraction time, average values for contraction length, peak to peak values, area under curve). Based on these measurements we determine the Spearman correlation coefficients and the scatter matrix between the FHR and UP signals and the fetal outcome parameters: pH and Apgar scores. The correlation result show a strong relationship between decelerations presented on the FHR signal, increasing number of contractions and the low pH and Apgar values.

[1]  J. Morrison.  Preterm Birth - Mother and Child , 2012 .

[2]  P. A. Warrick,et al.  Fetal heart-rate variability response to uterine contractions during labour and delivery , 2012, 2012 Computing in Cardiology.

[3]  Kazuo Maeda,et al.  Uterine Contractions in Normal Labor Developed by a Positive Feed-backand Oscillation , 2013 .

[4]  H. Costin,et al.  Automatic Assessing of Tremor Severity Using Nonlinear Dynamics, Artificial Neural Networks and Neuro-Fuzzy Classifier , 2014 .

[5]  E. M. Graatsma.  Monitoring of fetal heart rate and uterine activity , 2010 .

[6]  Janusz Wrobel,et al.  Analysis of extracted cardiotocographic signal features to improve automated prediction of fetal outcome , 2010 .

[7]  Antoniya Georgieva,et al.  Artificial neural networks applied to fetal monitoring in labour , 2011, Neural Computing and Applications.

[8]  B. B. Zaidan,et al.  Computerized Algorithm for Fetal Heart Rate Baseline and Baseline Variability Estimation based on Distance Between Signal Average and a Value , 2011 .

[9]  D. Bonett,et al.  Sample size requirements for estimating pearson, kendall and spearman correlations , 2000 .

[10]  Lenka Lhotská,et al.  Using nonlinear features for fetal heart rate classification , 2012, Biomed. Signal Process. Control..

[11]  J. Morrison,et al.  Uterine Contraction Monitoring, Maintenance Tocolysis, and Preterm Birth , 2012 .

[12]  Lenka Lhotská,et al.  Open access intrapartum CTG database , 2014, BMC Pregnancy and Childbirth.

[13]  R. González,et al.  Computerized algorithm for baseline estimation of fetal heart rate , 2002, Computers in Cardiology.

[14]  A. Fanaroff,et al.  Prediction of neonatal acidemia by computer analysis of fetal heart rate and ST event signals , 2010 .