Fetal ECG Separation from Abdominal ECG Recordings Using Compressive Sensing Approach
11 Jul 2018-pp 831-834
TL;DR: This paper is presenting the framework for detection of fetal from maternal ECG based on sparse binary matrix using Compressive Sensing, and presenting the preprocessing algorithm on raw fetus ECG data to remove the noises like impulsive artifacts along with notch filtering for baseline removal.
Abstract: The fetal electrocardiogram (f-ECG) beats analysis as well as the heart rate interpretation using the raw ECG signals captured by machine helps in providing the state of the fetus in pregnancy. For timely detection of the fetal arrhythmias, monitoring fetal ECGs constantly is important. In this paper, we are presenting the framework for detection of fetal from maternal ECG based on sparse binary matrix using Compressive Sensing. Additionally, we are presenting the preprocessing algorithm on raw fetus ECG data to remove the noises like impulsive artifacts along with notch filtering for baseline removal. The proposed method is on the basis of sparse representation of the components that are acquired using Independent Component Analysis (ICA) method, which is designed for direct application in the compressed domain. Detection of fetal ECG is performed on the basis of activated atoms in a specially designed Gaussian dictionary. The verification of the proposed framework has been carried out on ten samples of Challenge dataset A by determining QRS detection parameters such as sensitivity, $\mathbf{S}= 90.62\%$ and positive predictivity, given by P+= 99.15%.
Citations
More filters
18 Jul 2020
TL;DR: The present scheme uses one dimensional (1D) convolution with a wavelet kernel to extract time domain features from subjects possessing normal fetal ECG and fetal arrhythmia ECG to develop an intelligent system for portable embedded system applications.
Abstract: This paper aims to present an intelligent system for autonomous diagnosis of fetal arrhythmia based on fetal ECG recordings. The present scheme uses one dimensional (1D) convolution with a wavelet kernel to extract time domain features from subjects possessing normal fetal ECG and fetal arrhythmia ECG. Time- domain features obtained from the convoluted signals are fed to a trained artificial neural network (ANN) with gradient descent learning to identify and classify fetal ECG signals. The experimental evaluation of the proposed scheme has been tested with a six- channel fetal ECG signal, available in the NIFEADB database. An overall accuracy of 96% is obtained by evaluating standard performance metrics. The use of 1D convolution not only reduces the computational burden but also helps to specify the feature space to develop an intelligent system for portable embedded system applications.
5 citations
Additional excerpts
...ECG signal-based arrhythmia detection and classification has been reported in several literatures [5, 7-16]....
[...]
01 Jan 2021
TL;DR: In this paper, the authors proposed an early detection of fetal ECG and arrhythmia of the fetus using feature vector that is extracted from the signal and a classification method is also used to classify the signal as abnormal and normal.
Abstract: The infant mortality rate is the number of newborns death under 1 year of age occurring among the live births in a given region during a given year Electrocardiogram is generally used for finding the cardiovascular variation The infant mortality rate can be drastically reduced by adopting this proposed diagnosis technique for fetus This proposed work provides an indication of fetal health and heart information Sometimes newborn babies will be affected by heart diseases like tachycardia and bradycardia By diagnosing these diseases, we could reduce the death rate of the newborns This method proposes an early detection of fetal ECG and the arrhythmia of the fetus So this will dynamically reduce the infant mortality rate It brings amazing changes in the fields of medical industries and medical research fields The main aim of this work is to detect the variations in the heart rate The variations in the heart can be detected by using feature vector that is extracted from the signal and a classification method is also used to classify the signal as abnormal and normal This work shows a result of accuracy about 9411% and sensitivity of 8888%
1 citations
Patent•
19 Apr 2019
TL;DR: In this paper, a method for detecting the maternal electrocardiogram R peak on single-channel pregnant woman's abdominal wall myoelectricity signal is presented. But the method is characterized in that the Gaussian dictionary obtained in the step 3 is composed of Gaussian atoms corresponding to maternal EEG components, Gaussian nodes corresponding to fetal EEG components and Gaussian node corresponding to noise components.
Abstract: The invention provides a method for detecting the maternal electrocardiogram R peak on single-channel pregnant woman's abdominal wall myoelectricity. The method comprises the steps that 1, one-channelpregnant woman's abdominal wall myoelectricity signal is read; 2, the maternal electrocardiogram R peak on the abdominal wall myoelectricity signal is initially detected or pre-detected; 3, accordingto an initial detection result, an adaptive Gaussian dictionary is constructed, and based on the constructed adaptive Gaussian dictionary, enhancement of the maternal electrocardiogram R peak is achieved through sparse representation; 4, the R peak is detected on a maternal electrocardiogram R peak enhancement signal, and the position of the R peak is output. The method is characterized in that the Gaussian dictionary obtained in the step 3 is composed of Gaussian atoms corresponding to maternal electrocardiogram components, Gaussian atoms corresponding to fetal electrocardiogram components,and Gaussian atoms corresponding to noise components, wherein the Gaussian atoms corresponding to the maternal electrocardiogram components only involves one scale, and the scale is obtained through optimization according to the conditions of the maternal electrocardiogram R peak obtained through initial detection of the abdominal wall myoelectricity signal.
TL;DR: In this article , the authors presented the development of a belt composed of yarns containing different percentages of stainless-steel fibers as conductive yarns, which enabled the real-time monitoring of FHR by the mother in absence of medical staff.
Abstract: Fetal heart rate (FHR) monitoring is a technique that can be considered vital tool assisting midwife. This technique is useful in the early recognition of fetal heart defects, stillbirth and prevention of a risky conditions in clinics during pregnancy. The complications due to these threats are currently kept at bay by the extensive use of cardiotocography in clinics. However, cardiotocography suffers from drawbacks such as the need to replace transducer skillfully or the fact that the use of this method is only limited to mother to be staying in clinics. The aim of this article is to present the development of a belt composed of yarns containing different percentages of stainless-steel fibers as conductive yarns. The use of this belt in association with signal processing technique enables the real-time monitoring of FHR by the mother in absence of medical staff. Thus, not only preventing the occurrence of life threatening complexities but also significantly relieving the burden on the social welfare budget. It was concluded that a smart belt containing the highest percentage of stainless-steel fiber offers successful recording of fetal electrocardiogram.