Towards reconstruction of Pulsed-wave Doppler signals from Non-invasive fetal ECG
Aman Verma ∗, Deva Satya Sriram Chintapenta ∗, Saikat Majumder · 2025
Read Towards reconstruction of Pulsed-wave Doppler signals from Non-invasive fetal ECG by Aman Verma ∗, Deva Satya Sriram Chintapenta ∗, Saikat Majumder (2025) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2501-16651.txt.
Introduction
Aman Verma and two co-authors pose a hard question in fetal heart care in twenty twenty-five. Invasive monitoring of a baby's heart works only during labor and is uncomfortable. Non-invasive fetal ECG is safer but is swamped by the mother's own ECG signal. Pulsed-wave Doppler gives clean signals of fetal blood flow, yet its setup is very costly. The authors ask whether Doppler signals can be rebuilt from non-invasive fetal ECG instead. Their paper works toward that reconstruction, aiming for Doppler quality without Doppler hardware.
Catalog record
| CREATOR | Aman Verma ∗, Deva Satya Sriram Chintapenta ∗, Saikat Majumder |
|---|---|
| YEAR | 2025 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 26 KB |
| VERSION | 1.0 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2501-16651
Browse the full library · About the Tower of Babel
Full text
The opening of the work, from the library's hosted text:
Towards reconstruction of Pulsed-wave Doppler signals from Non-invasive fetal ECG Aman Verma ∗, Deva Satya Sriram Chintapenta ∗, Saikat Majumder Abstract Fetal cardiac health monitoring with invasive methods have a limited viability because they can only be utilized during labor and are uncomfortable. On the other hand non-invasive fECG are adulterated with maternal ECG, and hence resulting in poor analysis. In contrast, Pulsed-wave Doppler (PwD) echocardiography generates high-quality signals representing fetal blood volume inflow-outflow. It also follows non-invasive signal acquisition. The only drawback is that it requires highly expensive setup. To address this aspect, we put forward a challenging research question - can we reconstruct PwD signals using non-invasive fetal ECG? To answer this question, we perform a feasibility study with respect to input fECG wave polarity, output PwD signal configuration (EA+, EA- and group), output PwD envelopes (upper, lower, PCA compressed and group), signal length of input fECG signal and different regression models. In order to achieve good reconstruction, we also propose PwDRecNet – a deep learning framework which operates over multiple temporal contexts. To the best our knowledge this is the first work to consider PwD signal reconstruction from NI-fECG. The obtained numerical results suggests that with adequate configuration and model better reconstruction can be obtained. * * footnotetext: Equal contribution 1 Introduction For adults, Electrocardiogram (ECG) has been a gold-standard for cardiac activity monitoring, hence for arrhythmia prediction in fetus, fECG has been taken in account. fECG is defined as the electrical physiological signal generated by fetal cardiovascular system [ 5 , 6 , 7 ] . Scalp-invasive techniques have been used to measure it during labour, but this limits regular morphological fetal cardiac health monitoring. The method used to monitor fECG should both physically and economically feasible to be repeated as many numbers of times. Since invasive methods follow operation-based strategies they cannot be practiced at a prevalent scale. Non-invasive fECG (NI-fECG) has emerged as an alternative for invasive methods. It is collected via multi-electrode set-up placed over maternal abdomen. Although non-invasive, the NI-fECG signal comprises maternal ECG (MECG) and high noise [ 13 ] . This degrades the quality of analysis. Also, it does not encapsulates information about cardiac blood inflow and outflow. Figure 1: The figure illustrates NI-fECG and PwD signals. It also shows physiological relationship between ECG and PwD for an adult ((c) part of the figure is adapted from https://kenhub.com ). Pulsed-wave Doppler (PwD) signals on the other hand are more advanced techniques for cardiac health monitoring. It measures the blood velocity as well as the mitral inflow and aortic outflow, which conveys a richer physiological information with respect to fECG. Extraction of PwD signals involves ultrasound-based examination of fetus’s heart [ 9 ] . The spectral doppler signal does not only show blood velocity and corresponding flow volume, but also indicates the timing of cardiac events and intensity of flow (refer Fig. 1 (b)). Physiologically, PwD signal can be decomposed into atrioventricular cycles. Any such cycle can be characterized by three main waves of blood flow, when the blood flows through the mitral valve two waves are constituted: (i) E-Wave: This is the wave corresponding to passive filling of the ventricle due to the differential pressure between the two chambers, and (ii) A-Wave: This is the wave corresponding to the active filling of ventricle due to atrial contraction. These two waves represent the atrial blood inflow activity. In contrast, the aortic outflow is represented via the V-Wave . Although, the signal encompasses significant information regarding fetal cardiac cycles, it’s acquisition requires an expensive setup. This restricts its wide-scale usage, especially in low-income countries. To this end, we find consistent patterns between ECG and ventricular volume, as illustrated in Fig. 1 (c). It is evident from the same figure that ECG uniformly changes with respect to ventricular volumetric flow of blood. This corresponds to the fact that there exists morphological correspondence between EAV-Waves and respective ECG signals. Being motivated from the same, we put forward a research question: "Can we reconstruct the PwD signal using the Non-invasive fECG signal?” By realizing this reconstruction, affordable and convenient NI-fECG signals can utilized to recover highly informative PwD signal without needing any specific setup. In order to explore the feasibility of the same, we experiment with multiple physiological settings. The results obtained are encouraging, but suggests there is requirement of lot of research thrust to enable such cross-modality reconstruction. Following are the key contributions of this work: • We propose PwD signal envelope reconstruction using ubiquitous NI-fECG signals. To the best of our knowledge, this is the first work to explore this aspect. • In order to check upon the feasibility of the reconstruction, we conduct in-depth validation on output wave configurations, fECG polarities, timing of samples, and different models. • We also propose PwDRecNet, a residually connected UNet style architecture which capture contexts at multiple-levels. In the next section we present a literature review and then in the following section methodology is explained. Then, in subsequent section experimental Analysis is presented, while the report is concluded in the section that follows. 2 Related works 2.1 PwD signal reconstruction As mentioned earlier, it is challenging to extract discriminative features from the NI-fECG signal due to inherent noise and multi-signal waveform. Hence, it has been a key research question – ‘How to separate fECG from MECG?’. To this end, research in [ 4 , 13 ] utilized Attention-Mechanism along with CycleGAN to achieve the state-of-the-art results. As mentioned earlier, NI-fECG does not reveal much information about blood dynamics inside the fetal heart, which in-turn limits the diagnosis of anomaly localization. Hence, usage of PwD based fetal cardiac health diagnosis seems a brighter choice. Towards, this in [ 9 ] authors tried to classify the cycles of PwD signals in EA+, EA- and incomplete cycle. Towards the same, authors extracted the envelopes and then used an artificial neural network for classification. The achieves performance was satisfactory but left open research gaps. In the subsequent work [ 10 ] , the same authors increased the dataset size and utilized similar method to validate on larger scales. Finally, in [ 11 ] authors gave three different approaches to classify the PwD envelopes as full cycles or incomplete cycles. They utilized fiducial methods as well as ANN based methods to realize the task. Although, there had been a recent thrust towards PwD, the field remains significantly open. There had been no methods to extract PwD signals in more economical manner, towards this end, this project presents a novel approach to alleviate the same. 2.2 Cross-modality biosignal reconstruction Bio-signal reconstruction has recently caught attention, in a pilot study, authors in [ 14 ] tried to reconstruct ECG from more ubiquitous signal – PPG. The results obtained established a strong baseline and further motivated physiology to physiology domain adaption. In [ 15 ] , authors further improvised the ECG reconstruction using PPG, they also gave a mathematical model and utilized different regression algorithms. Similar to ECG reconstruction, in [ 16 ] pulsed-wave and respiration were reconstructed using wrist accelerometer data. Authors used primitive statistical models for the reconstruction and hence there remains a gap to further improvise the performance using more sophisticated