Estimation of Fetal Autonomic Age By Considering Fetal Behavioural States

Prior research has demonstrated that maternal heart rate variability (HRV) and physiological factors can significantly impact fetal autonomic nervous system development. This article aims to predict fetal autonomic age using the multi-linear regression models constructed from different features including maternal HRV, respiration rate, and demographics, along with maternal-fetal heartbeats coupling strength in addition to fetal HRV parameters. The study includes fetal ECG data of 62 fetuses with gestational ages ranging between 20–41 weeks. Initially, the data is segmented according to fetal behavioural states and then used in the proposed methodology. Leave one sample out cross-validation method is used to validate the models. Results showed the importance of considering maternal HRV parameters, BMI, and respiration rate when estimating fetal age in active state, where the proposed methodology returned a correlation of 0.604 and error of 2.95 weeks. In quiet state, more involvement of maternal-fetal heartbeats coupling strength was observed when estimating the fetal age (R² = 0.75, Error = 2.86 weeks). Therefore, the findings presented in this study confirmed that the assessment of fetal development and health can be improved by incorporating maternal features.