Journal
IEEE ACCESS
Volume 9, Issue -, Pages 65369-65379Publisher
IEEE-INST ELECTRICAL ELECTRONICS ENGINEERS INC
DOI: 10.1109/ACCESS.2021.3074550
Keywords
Biomedical signal processing; electrocardiography; fetal development; fetal heart rate; gestational age; heart rate variability; linear models; pregnant women
Categories
Funding
- Khalifa University, Abu Dhabi, UAE [CIRA 2019-023, 8474000174]
Ask authors/readers for more resources
This study aimed to investigate the association between maternal-fetal heart rate coupling parameters and fetal development, as well as to estimate fetal gestational age using these parameters. By considering heart rate coupling strength and variability across different recording lengths, the study found that a multivariate model incorporating various parameters can more reliably estimate fetal gestational age.
Maternal and fetal heartbeat couplings are evident throughout fetal development. Most of the published work, however, did not consider maternal physiological factors such as Heart Rate Variability (HRV), and did not investigate the interrelationships of maternal-fetal coupling parameters. The aims of this study are to investigate whether: 1) maternal-fetal Heart Rate (HR) coupling (lambda-based) parameters are associated with fetal development, and 2) fetal gold standard Gestational Age (GA) can be estimated using maternal-fetal HR coupling and variability of various recording lengths. The study considered Electrocardiogram (ECG) signals from 60 healthy pregnant women with no records of fetal abnormalities. HRV and fi parameters at various Maternal:Fetal coupling ratios were calculated, and stepwise regression was utilized to create generalized linear regression models considering various lengths of recorded signals (1 and 5 min) to produce a robust estimate of fetal age. Cross-validation performances were evaluated by the mean square root of the average of squared errors (mRMSE) between age values estimated by the proposed models and gold standard GA identified by Crown-Rump Length (CRL). Effect of Fetal Behavioral States (FBSes) on proposed models with different recording lengths was considered to examine the highly nonstationary nature of signals. We found that HR coupling strength for a specific ratio is not constant throughout gestation. Results showed that ratios of 2:3 and 2:4 were common between the proposed models. The value of lambda[2:3] was found to be positively correlated with GA, while lambda[2:4] had a negative correlation. Compared with gold standard GA identified by CRL, the proposed regression model resulted in mRMSE of 2.67 and 3.69 weeks for the recordings of 5 and 1 min, respectively. However, when FBS was considered, both models produced lower estimation errors. Fetal GA can be more reliably estimated by a multivariate model incorporating fetal and maternal HR coupling and HRV parameters using 5 min of ECG signals.
Authors
I am an author on this paper
Click your name to claim this paper and add it to your profile.
Reviews
Recommended
No Data Available