AECG-DecompNet: abdominal ECG signal decomposition through deep-learning model.

Researchers

Journal

Modalities

Models

Abstract

An accurate decomposition of abdominal electrocardiogram (AECG) of the mother to extract Fetal ECG (FECG) is a primary step to evaluate the fetus’s health. However, AECG is often affected by different noises and interferences, such as Maternal ECG (MECG), making it hard to evaluate the FECG signal. In this paper, we propose a deep learning-based framework, namely \AECG-DecompNet”, to efficiently extract both MECG and FECG from a single-channel abdominal electrode recording.
The AECG-DecompNet is based on two series networks to decompose AECG, one for MECG estimation and the other to eliminate interference and noise. Both networks are based on an encoder-decoder architecture with internal and external skip connections to reconstruct the signals better.
Experimental results show that the proposed framework performs much better than utilizing one network for direct FECG extraction. In addition, the comparison of the proposed framework with popular single-channel extraction techniques shows superior results in terms of QRS detection while indicating to be able to preserve morphological information. AECG-DecompNet achieves an exceptional accuracy in the P recision metric (97.4%) and higher accuracy in Recall and F1 metrics (93.52% and 95.42% respectively) and outperforms other state-of-the-art approaches.
The proposed method shows a notable performance in preserving the morphological information when the FECG within the AECG signal is weak.
© 2021 Institute of Physics and Engineering in Medicine.