An Efficient and Robust Deep Learning Method with 1-D Octave Convolution to Extract Fetal Electrocardiogram

Abstract

<jats:p>The invasive method of fetal electrocardiogram (fECG) monitoring is widely used with electrodes directly attached to the fetal scalp. There are potential risks such as infection and, thus, it is usually carried out during labor in rare cases. Recent advances in electronics and technologies have enabled fECG monitoring from the early stages of pregnancy through fECG extraction from the combined fetal/maternal ECG (f/mECG) signal recorded non-invasively in the abdominal area of the mother. However, cumbersome algorithms that require the reference maternal ECG as well as heavy feature crafting makes out-of-clinics fECG monitoring in daily life not yet feasible. To address these challenges, we proposed a pure end-to-end deep learning model to detect fetal QRS complexes (i.e., the main spikes observed on a fetal ECG waveform). Additionally, the model has the residual network (ResNet) architecture that adopts the novel 1-D octave convolution (OctConv) for learning multiple temporal frequency features, which in turn reduce memory and computational cost. Importantly, the model is capable of highlighting the contribution of regions that are more prominent for the detection. To evaluate our approach, data from the PhysioNet 2013 Challenge with labeled QRS complex annotations were used in the original form, and the data were then modified with Gaussian and motion noise, mimicking real-world scenarios. The model can achieve a F1 score of 91.1% while being able to save more than 50% computing cost with less than 2% performance degradation, demonstrating the effectiveness of our method.</jats:p>

Deep Learning-Based Computer-Aided Fetal Echocardiography: Application to Heart Standard View Segmentation for Congenital Heart Defects Detection

“…In this study, the instance segmentation approach is developed based on Mask-RCNN architecture (refer to Figure 6 ) [ 18 , 22 ]. The Mask-RCNN structure has two main processes, region proposal networks (RPNs) as feature extraction and fully convolutional networks (FCNs) as multi-task learning process in terms of simultaneous classification, detection, and segmentation.…”

“…These conditions are very dangerous, as they allow shunt of blood flow from the right heart chambers to the left, and vice versa [ 3 ]. The deep learning (DL)-based convolutional neural networks (CNNs) architecture is an AI approach that can apply to fetal object diagnosis [ 6 , 7 , 9 , 10 , 11 , 12 , 13 , 14 , 15 , 16 , 17 , 18 ].…”

“…Several studies have reported powerful results regarding CNNs’ ability in segmentation, classification, and detection based on medical imaging [ 15 , 16 , 17 , 18 , 19 , 20 , 21 ]. CNNs, which are applications that perform adaptation functions without being specifically programmed, learn from data and make accurate predictions or decisions based on past data [ 6 , 7 , 8 , 9 , 10 ].…”

A Multilayer LSTM Auto-Encoder for Fetal ECG Anomaly Detection

“…The results show that the error vectors represented by their summary features carry useful predictive information about the type of the ECG anomaly. In [7], a deep learning model based on residual network (ResNet) that adopts the 1-D octave convolution was proposed to extract fetal ECG. The study [8] proposed a cardiac abnormality detection algorithm for optical photoplethysmography (PPG) based on an autoencoder.…”

Modern Trends and Applications of Intelligent Methods in Biomedical Signal and Image Processing

“…Recent advances in electronics and technologies have enabled fECG monitoring from the early stages of pregnancy through fECG extraction from the combined fetal/maternal ECG (f/mECG) signal recorded noninvasively in the abdominal area of the mother. In Reference [ 6 ], the authors propose an end-to-end deep learning model which is aimed at the detection of fetal QRS complexes. The proposed model also contains the residual network (resNet) architecture.…”