A Complex UNet Approach for Non-Invasive Fetal ECG Extraction Using Single-Channel Dry Textile Electrodes

• URL: http://arxiv.org/abs/2506.22457v1
• Date: Mon, 16 Jun 2025 20:35:52 GMT
• Title: A Complex UNet Approach for Non-Invasive Fetal ECG Extraction Using Single-Channel Dry Textile Electrodes
• Authors: Iulia Orvas, Andrei Radu, Alessandra Galli, Ana Neacsu, Elisabetta Peri,
• Abstract summary: The fetal electrocardiogram (fECG) represents a promising tool for assessing fetal health beyond clinical environments.<n>This setup presents many challenges, including increased noise and motion artefacts, which complicate the accurate extraction of fECG signals.<n>We introduce a pioneering method for extracting fECG from single-channel recordings obtained using dry textile electrodes using AI techniques.
• Score: 40.11095094521714
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Continuous, non-invasive pregnancy monitoring is crucial for minimising potential complications. The fetal electrocardiogram (fECG) represents a promising tool for assessing fetal health beyond clinical environments. Home-based monitoring necessitates the use of a minimal number of comfortable and durable electrodes, such as dry textile electrodes. However, this setup presents many challenges, including increased noise and motion artefacts, which complicate the accurate extraction of fECG signals. To overcome these challenges, we introduce a pioneering method for extracting fECG from single-channel recordings obtained using dry textile electrodes using AI techniques. We created a new dataset by simulating abdominal recordings, including noise closely resembling real-world characteristics of in-vivo recordings through dry textile electrodes, alongside mECG and fECG. To ensure the reliability of the extracted fECG, we propose an innovative pipeline based on a complex-valued denoising network, Complex UNet. Unlike previous approaches that focused solely on signal magnitude, our method processes both real and imaginary components of the spectrogram, addressing phase information and preventing incongruous predictions. We evaluated our novel pipeline against traditional, well-established approaches, on both simulated and real data in terms of fECG extraction and R-peak detection. The results showcase that our suggested method achieves new state-of-the-art results, enabling an accurate extraction of fECG morphology across all evaluated settings. This method is the first to effectively extract fECG signals from single-channel recordings using dry textile electrodes, making a significant advancement towards a fully non-invasive and self-administered fECG extraction solution.

Related papers

• Diffusion-Based Electrocardiography Noise Quantification via Anomaly Detection [2.741077302469742]
  Existing methods for addressing ECG noise, relying on artifact classification or denoising, are constrained by annotation inconsistencies and poor generalizability.<n>We propose a diffusion-based framework trained to model the normative distribution of clean ECG signals, identifying deviations as noise without requiring explicit artifact labels.
  arXiv  Detail & Related papers  (2025-06-13T14:19:04Z)
• SincVAE: a New Approach to Improve Anomaly Detection on EEG Data Using SincNet and Variational Autoencoder [0.0]
  This work proposes a semi-supervised approach for detecting epileptic seizures from EEG data, utilizing a novel Deep Learning-based method called SincVAE. Results indicate that SincVAE improves seizure detection in EEG data and is capable of identifying early seizures during the preictal stage as well as monitoring patients throughout the postictal stage.
  arXiv  Detail & Related papers  (2024-06-25T13:21:01Z)
• CathFlow: Self-Supervised Segmentation of Catheters in Interventional Ultrasound Using Optical Flow and Transformers [66.15847237150909]
  We introduce a self-supervised deep learning architecture to segment catheters in longitudinal ultrasound images. The network architecture builds upon AiAReSeg, a segmentation transformer built with the Attention in Attention mechanism. We validated our model on a test dataset, consisting of unseen synthetic data and images collected from silicon aorta phantoms.
  arXiv  Detail & Related papers  (2024-03-21T15:13:36Z)
• EKGNet: A 10.96{\mu}W Fully Analog Neural Network for Intra-Patient Arrhythmia Classification [79.7946379395238]
  We present an integrated approach by combining analog computing and deep learning for electrocardiogram (ECG) arrhythmia classification. We propose EKGNet, a hardware-efficient and fully analog arrhythmia classification architecture that archives high accuracy with low power consumption.
  arXiv  Detail & Related papers  (2023-10-24T02:37:49Z)
• AiAReSeg: Catheter Detection and Segmentation in Interventional Ultrasound using Transformers [75.20925220246689]
  endovascular surgeries are performed using the golden standard of Fluoroscopy, which uses ionising radiation to visualise catheters and vasculature. This work proposes a solution using an adaptation of a state-of-the-art machine learning transformer architecture to detect and segment catheters in axial interventional Ultrasound image sequences.
  arXiv  Detail & Related papers  (2023-09-25T19:34:12Z)
• DGSD: Dynamical Graph Self-Distillation for EEG-Based Auditory Spatial Attention Detection [49.196182908826565]
  Auditory Attention Detection (AAD) aims to detect target speaker from brain signals in a multi-speaker environment. Current approaches primarily rely on traditional convolutional neural network designed for processing Euclidean data like images. This paper proposes a dynamical graph self-distillation (DGSD) approach for AAD, which does not require speech stimuli as input.
  arXiv  Detail & Related papers  (2023-09-07T13:43:46Z)
• ECG Artifact Removal from Single-Channel Surface EMG Using Fully Convolutional Networks [9.468136300919062]
  This study proposed a novel denoising method to eliminate ECG artifacts from the single-channel sEMG signals using fully convolutional networks (FCN) The proposed method adopts a denoise autoencoder structure and powerful nonlinear mapping capability of neural networks for sEMG denoising.
  arXiv  Detail & Related papers  (2022-10-24T14:12:11Z)
• fMRI from EEG is only Deep Learning away: the use of interpretable DL to unravel EEG-fMRI relationships [68.8204255655161]
  We present an interpretable domain grounded solution to recover the activity of several subcortical regions from multichannel EEG data. We recover individual spatial and time-frequency patterns of scalp EEG predictive of the hemodynamic signal in the subcortical nuclei.
  arXiv  Detail & Related papers  (2022-10-23T15:11:37Z)
• Grasp-and-Lift Detection from EEG Signal Using Convolutional Neural Network [1.869097450593631]
  This article has automated the hand movement activity viz GAL detection method from the 32-channel EEG signals. The proposed pipeline essentially combines preprocessing and end-to-end detection steps, eliminating the requirement of hand-crafted feature engineering.
  arXiv  Detail & Related papers  (2022-02-12T19:27:06Z)
• Representing and Denoising Wearable ECG Recordings [12.378631176671773]
  We develop a statistical model to simulate a structured noise process in ECGs derived from a wearable sensor. We design a beat-to-beat representation that is conducive for analyzing variation, and devise a factor analysis-based method to denoise the ECG.
  arXiv  Detail & Related papers  (2020-11-30T21:33:11Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.