Deep learning identifies cardiac coupling between mother and fetus during gestation

Abstract

In the last two decades, stillbirth has caused around 2 million fetal deaths worldwide. Although current ultrasound tools are reliably used for the assessment of fetal growth during pregnancy, it still raises safety issues on the fetus, requires skilled providers, and has economic concerns in less developed countries. Here, we propose deep coherence, a novel artificial intelligence (AI) approach that relies on 1 min non-invasive electrocardiography (ECG) to explain the association between maternal and fetal heartbeats during pregnancy. We validated the performance of this approach using a trained deep learning tool on a total of 941 one minute maternal-fetal R-peaks segments collected from 172 pregnant women (20–40 weeks). The high accuracy achieved by the tool (90%) in identifying coupling scenarios demonstrated the potential of using AI as a monitoring tool for frequent evaluation of fetal development. The interpretability of deep learning was significant in explaining synchronization mechanisms between the maternal and fetal heartbeats. This study could potentially pave the way toward the integration of automated deep learning tools in clinical practice to provide timely and continuous fetal monitoring while reducing triage, side-effects, and costs associated with current clinical devices.

Evidence and clinical relevance of maternal-fetal cardiac coupling: A scoping review

“…In other words, it does not say whether the fetal HR affects the maternal HR or the other way around. Additionally, the final study included in Table 1 uses an artificial intelligence method known as Deep coherence [ 37 ]. This method is a deep learning implementation of the phase coherence index, where the deep learning model seeks to identify phases of synchronization in correspondence with what would be found with the original method described above, but without any mathematical derivations, pre-processing steps, or signal transformations to the input data [ 37 ].…”

“…Lastly, a deep learning approach called deep coherence was recently proposed in the field of MFCC research [ 37 ]. Deep learning methods like deep coherence may help to reduce the need for a priori assumptions and processing.…”

Gender of Fetus Identification Using Modified Mel-Frequency Cepstral Coefficients Based on Fractional Discrete Cosine Transform

“…In 2022, M. Farahi et al used portable device and wavelet transform to made heart rate fetal analysis [8]. In 2022, M. Alkhodari et al used deep learning to identify cardiac coupling between mother and fetus during gestation [9]. In 2023, R. Abburi and others used artificial intelligence algorithms for remote fetal heart rate monitoring and classification [10].…”