Corrupted ECG Signal Research Articles

FPGA implementation of IIR elliptic filters for de-noising ECG signal

FPGA implementation of IIR elliptic filters for de-noising ECG signal

Transfer learning based deep network for signal restoration and rhythm analysis during cardiopulmonary resuscitation using only the ECG waveform

Minimizing the interruption of cardiopulmonary resuscitation (CPR) is an important technique to improve the survival of out-of-hospital cardiac arrest (OHCA) patients. Recent studies have adopted deep convolutional neural networks (CNNs) to analyze corrupted ECGs during CPR for shockable rhythm detection, however, none of them provided restored ECG signals using only the ECG waveform. In this study, a self-supervised UNet deep learning network is designed to restore the underlying ECG signals during chest compressions without additional reference signals. The UNet is pretrained using a large dataset created by combining 12-lead clean ECG signals with simulated CPR artifacts generated by a mathematical model. Transfer learning is applied to fine-tune the UNet parameters based on combined ECG signals from OHCA patients. Finally, an independent dataset extracted from OHCA patients is used to test the model. Compared with the artifact corrupted ECG signals, the overall signal-to-noise ratio (SNR) is improved from -5.3 (-11.2, 0.8) dB to 1.9 (0.7, 3.9) dB after the restoration, and the accuracy of the rhythm detection is improved from 65.8% to 90.8%. The proposed method not only effectively restores the underlying ECG signals but also provides accurate shock advice without interrupting CPR using only the ECG.

Blind ECG Restoration by Operational Cycle-GANs.

ECG recordings often suffer from a set of artifacts with varying types, severities, and durations, and this makes an accurate diagnosis by machines or medical doctors difficult and unreliable. Numerous studies have proposed ECG denoising; however, they naturally fail to restore the actual ECG signal corrupted with such artifacts due to their simple and naive noise model. In this pilot study, we propose a novel approach for blind ECG restoration using cycle-consistent generative adversarial networks (Cycle-GANs) where the quality of the signal can be improved to a clinical level ECG regardless of the type and severity of the artifacts corrupting the signal. To further boost the restoration performance, we propose 1D operational Cycle-GANs with the generative neuron model. The proposed approach has been evaluated extensively using one of the largest benchmark ECG datasets from the China Physiological Signal Challenge (CPSC-2020) with more than one million beats. Besides the quantitative and qualitative evaluations, a group of cardiologists performed medical evaluations to validate the quality and usability of the restored ECG, especially for an accurate arrhythmia diagnosis. As a pioneer study in ECG restoration, the corrupted ECG signals can be restored to clinical level quality. By means of the proposed ECG restoration, the ECG diagnosis accuracy and performance can significantly improve.

FPGA-based reservoir computing system for ECG denoising

• SNRC is proposed to overcome noises that corrupt ECG signals. • Two typical types of ECG-contaminating noises are simulated and tested. • The proposed SNRC affords better results than the previous techniques. • An FPGA-based SNRC architecture is implemented to afford high speed and privacy. • The floating point (FP) arithmetic is used to obtain more accurate results. ECG is corrupted by various noises such as baseline wander, channel noise, Electromyogram (EMG) noise, and power line interference (PLI). These noises make diagnosis difficult. In this paper, we implement an inexpensive, portable device that can remove noise from the corrupted ECG signals with high performance, speed, and privacy. First, we apply a Single Node Reservoir Computing (SNRC) architecture to clean the corrupted ECG signal with high performance. Second, we implement our technique on a portable inexpensive FPGA device that enables us to achieve high speed and privacy. In this manuscript, we simulate two noises: the typical EMG and PLI noises. To evaluate our technique, we use three performance metrics, namely, the output SNR improvement (SNRimp), the mean square error (MSE), and the percentage root mean square difference (PRD). The data is collected from the Massachusetts Institute of Technology-Boston's Beth Israel Hospital (MIT-BIH) arrhythmia database. With input SNR = 0 dB, the proposed system achieves SNRimp of 15.8 and PRD of 24.6, in the case of EMG noise and SNRimp of 25.7and PRD of 4.9, in the case of PLI noise. Comparison results to the related works show the advantages of the proposed system.

New approach to eliminate noise attendants ECG signal corrupted

ECG signals have been largely utilized in cardiac pathology to detect heart ailments. ECG signals are commonly corrupted by disparate artifacts such as power line obtrusion exterior electromagnetic domain intervention, noise due to random object motions, breathing motions, and electromyography noise and these should be eliminated before diagnosis.This article explains the creation and performance of three types of band-stop filters based on FPAA technology to accurately estimate an ECG corrupted signal. The characteristics of these filters are improved and their parameters are calculated to minimize the corrupted ECG signals with noise. The Butterworth filter of the fourth order, Notch filters, and least-mean-square adaptive filter (LMSAF) stop-band are used to reduce noise interference from ECG signals. The performance of filters was analyzed by comparing signal power at 62 Hz before and after filtration and distortion to the ECG waveform. It is found that digital filters work satisfactorily. LMSAF gives the best performance as the QRST segment of the ECG signal is less modified compared to others

A Novel Approach of ECG Signal Enhancement Using Adaptive Filter Based on Whale Optimization Algorithm

An Electro cardiogram is commonly used in biomedical signal processing. It is used to monitor minor electrical changes in the human body. The electrical changes originate due to the function of heart. The anomalies of heart are found by ECG. In this work the Whale optimization algorithm is used to de-noising the ECG signal. The Whale optimization algorithm is used with Adaptive filter which filter the corrupted ECG signal. The performance of the ANC will be improved by calculating the optimum weight value. The WOA technique gives the best result on the different fidelity parameter compare to PSO, MPSO and ABC. The WOA technique gives the significant improvement in accuracy. It gives a good SNR, MSE, ME result compare to PSO, MPSO and ABC. The WOA gives 80% improvement in SNR 88% in MSE and 89% in ME as compared to PSO. So, by using WOA we get a desired ECG component. The WOA reduces the noise in ECG signal and improves the quality of signal.

Synchrosqueezing Transform Based Powerline Interference Reduction in ECG Recording

This paper presents an adaptive ECG enhancement procedure based on Synchrosqueezing Transform (SST) to eliminate Powerline interference (PLI) from ECG signal. This work also incorporates the principles of modified discrete cosine transform (MDCT) and wiener filter. PLI is a major source of artifacts in the ECG signal which can affect its interpretation. Separating PLI from ECG signal poses a great challenge in the ECG analysis. The existing PLI removal techniques suffer from two major drawbacks such as Mode Mixing, inability to deal with non-stationary characteristics of signal. In this paper, we propose SST based wiener filtering approaches which can overcome the limitation of existing PLI suppression techniques. The proposed approaches undergo three stages of operation: mode decomposition, mode determination and peak restoration to filter out PLI from ECG recording. The mode decomposition uses SST to decompose the corrupted ECG signal into a sum of well separated intrinsic mode functions (IMFs). The objective is to filter out PLI from these IMFs. To do so, mode determination step which is based on Kurtosis and Crest factor is applied to categorize decomposition result into groups such as signal mode and noisy mode. Direct subtraction of the noisy mode from the corrupted ECG observation results in ECG signal with reduced peak since noise mode carries part of signal components in addition to interference. Hence, to restore the peak, wiener filter is applied on noisy mode to estimate actual PLI component. Finally, Noise free ECG signal is reconstructed by subtracting estimated PLI from the corrupted ECG signal. This paper discusses four possible PLI suppression methods which are derived by combining SST domain with wiener filter in various ways. Simulations are carried out to test the effectiveness of proposed methods. It is evident from the simulation results that the proposed methods can remove PLI of 50 Hz and its harmonics. The proposed techniques effectively removed PLI in both real and artificial ECG signals and to test its performance they are compared with state of the art methods. The SST based filtering methods outperformed other methods under the condition of PLI frequency variations. The experimental results also suggest that the SST based wiener filtering with modified reference approach offers better PLI suppression than all other methods.

Revisiting signal processing with spectrogram analysis on EEG, ECG and speech signals

Abstract Biomedical signal processing is the utilization of digital signal processing techniques, such as Fourier transform, filtering, spectral estimation and wavelet transform to biomedical complications, such as the analysis of breathing cycle, cardiac signals, brain signals, etc. Digital filters are used to preserve the in-band signals and to block out-of-band noise. Low-pass, high-pass, band-pass, and band-stop filters are commonly used for filtering applications. The digital signal processing (DSP) concepts can be used for other biomedical applications such as biomedical imaging (MRI, CT, X-ray, PET, ultrasound) and genomic signal processing too. The experiments such as denoising corrupted ECG signals, respiratory artifacts removal, and an identification of rhythmic patterns from EEG signals with spectrogram analysis are performed. Matlab R2016b tool is used for this study. Finally, the assessment, survey and feedback results from the student are tabulated for the better improvement of the course study.

Adaptive Artifact Cancelation Based on Bacteria Foraging Optimization for ECG Signal

In this paper, the design of adaptive artifact canceler (AAC) filter using bacteria foraging optimization (BFO) algorithm is presented. The performance of proposed AAC filter is tested on a corrupted ECG signal. Based on simulation results, it is observed that the AAC filter designed with BFO technique achieves significant improvement in fidelity parameters such as SNR, NRMSE, and NRME when compared with other reported algorithms in the literature. AAC filter based on BFO technique provides 6 dB improvement in output SNR, 85% reduction in NRMSE, and 90% lower NRME as compared to recently reported AAC filter based on ABC-SF algorithm. Further, AAC filter using BFO technique enhances the coherence between pure and reconstructed ECG signals.

A Study of Sensitivity of a Least Mean Square Filter on its Tap Weight Vector Length and Step Size in Adaptively Cancelling Noise in an Electrocardiogram Signal

m (ECG) is a procedure that records the activity of the heart and presents it as an electrical signal. It is an immensely important diagnostic tool since any deviation from the characteristic shape of an ECG may point to a cardiac anomaly, making it important for instrument and processing to be very accurate. Since the nature of noise that could corrupt the ECG is non-stationary, Adaptive Noise Cancellation (ANC) filters are required. In this paper, the Adaptive Noise Cancellation algorithm Least Mean Squares (LMS) has been implemented to reduce the noise in a corrupted ECG signal. The parameter sensitivity of the filter on its result had been studied, how the values of tap weight vector length and step size of the filter influence the quality of the resultant signal of the filter. This has been done through simulations in MATLAB, the noise have been simulated in MATLAB and the ECG signals have been collected from the ECG-ID database at PhysioNet. For various values of tap weight vector length and step size the performance of the LMS filter is analyzed in terms of PRD and MSE and the observations are tabulated. At the end of this study parameter values are suggested that render the most optimum results.

An Enhanced Adaptive Filtering Method for Suppressing Cardiopulmonary Resuscitation Artifact.

Cardiopulmonary resuscitation (CPR) must be interrupted for reliable rhythm analysis in current automatic external defibrillators because of artifacts produced by chest compressions. However, interruptions in CPR adversely affect the restoration of spontaneous circulation and survival. Suppressing CPR artifacts by digital signal processing techniques is a promising method to enable rhythm analysis during chest compressions, which would eliminate CPR interruptions for rhythm analysis. Although numerous methods have been developed to suppress CPR artifacts, the accuracy of rhythm analysis is still inadequate due to the residual artifact components in the filtered signal. This study proposes an enhanced adaptive filtering method to suppress CPR artifacts. A total of 183 shockable and 453 nonshockable segments of ECG signal, together with CPR-related reference signal, were extracted from 233 out of hospital cardiac arrest patients. The method was optimized on a training set with 85 shockable and 211 nonshockable segments, and evaluated on a testing set with 98 shockable and 242 nonshockable segments. Compared with artifact corrupted ECG signals, the signal-to-noise ratio (SNR) increased from -9.8±12.5 to 11.2±11.8dB, and the accuracy was improved from 74.1% to 92.0% after filtering with the proposed method. Compared with the traditional adaptive filter, the SNR was improved by 1.7dB and the accuracy was improved by 5.6 points. These results indicated that the proposed method could effectively suppress the chest compression related artifacts and improve the accuracy of rhythm analysis during uninterrupted CPR.

Adaptive filtering using PSO, MPSO and ABC algorithms for ECG signal

In this paper, the design of Adaptive Noise Canceller (ANC) filter using evolutionary algorithms such as Particle Swarm Optimisation (PSO), Modified PSO (MPSO) and Artificial Bee Colony (ABC) algorithms is presented. The performance of proposed ANC filter is tested on a corrupted ECG signal. Based on simulation results, it is observed that the ANC filter constructed using these evolutionary algorithms achieves significant improvement in fidelity parameters such as SNR, MSE, ME and correlation factor when compared with other reported techniques in literature. ANC filter based on ABC with scaling factor provides 78% improvement in output SNR, 76% and 87% reduction in MSE and ME, respectively, as compared to ANC filter based on PSO. Further, ANC filter designed using ABC technique enhances the correlation between output and pure ECG signal.

SWARM ALGORITHM BASED ADAPTIVE FILTER DESIGN TO REMOVE POWER LINE INTERFERENCE FROM ECG SIGNAL

ECG signal is having wide importance in the biomedical field, but for proper diagnosis of ECG always a noise free ECG signal is needed. Many researchers have already developed filters for getting appropriate desirable ECG signal and till today many researchers are still developing different filters using different algorithms in order to get clearer ECG signal for proper diagnosis. Noises and Interferences get added in the ECG by different ways, at the time of ECG Acquisition or at the time of ECG signal recording. In this paper newly adapted algorithm is used for the filtering of ECG signal that is a Swarm algorithm which is used for the Error signal optimization from the original corrupted ECG signal. This algorithm is implemented with Adaptive filter to removes Power Line Interference noise having Frequency component of 50 Hz. The ECG signal considered may be retrieved from ECG acquisition system or from MIT-BIH database.

Removal of Cardiopulmonary Resuscitation Artifacts with an Enhanced Adaptive Filtering Method: An Experimental Trial

Current automated external defibrillators mandate interruptions of chest compression to avoid the effect of artifacts produced by CPR for reliable rhythm analyses. But even seconds of interruption of chest compression during CPR adversely affects the rate of restoration of spontaneous circulation and survival. Numerous digital signal processing techniques have been developed to remove the artifacts or interpret the corrupted ECG with promising result, but the performance is still inadequate, especially for nonshockable rhythms. In the present study, we suppressed the CPR artifacts with an enhanced adaptive filtering method. The performance of the method was evaluated by comparing the sensitivity and specificity for shockable rhythm detection before and after filtering the CPR corrupted ECG signals. The dataset comprised 283 segments of shockable and 280 segments of nonshockable ECG signals during CPR recorded from 22 adult pigs that experienced prolonged cardiac arrest. For the unfiltered signals, the sensitivity and specificity were 99.3% and 46.8%, respectively. After filtering, a sensitivity of 93.3% and a specificity of 96.0% were achieved. This animal trial demonstrated that the enhanced adaptive filtering method could significantly improve the detection of nonshockable rhythms without compromising the ability to detect a shockable rhythm during uninterrupted CPR.

De-Noising Corrupted ECG Signals By Empirical Mode Decomposition (EMD) With Application of Higher Order Statistics (HOS)

De-Noising Corrupted ECG Signals By Empirical Mode Decomposition (EMD) With Application of Higher Order Statistics (HOS)

Abstract 271: Detection of Ventricular Fibrillation with Continuing Chest Compression Using Adaptive Least Mean-Square Estimation of Cardiopulmonary Resuscitation Artifacts and Removal of the Artifacts

Introduction: Adequate perfusion during cardiopulmonary resuscitation (CPR) is crucial for successful defibrillation after ventricular fibrillation (VF), but the effects of chest compressions (CC) are compromised from interruptions due to rhythm analysis. A novel algorithm utilizing corrupted ECG signal to detect VF during continuing CC is developed. Method: VF waveforms were retrieved from automatic external defibrillator (AED) records of adult non-traumatic cardiac arrests. Ten CC artifacts were retrieved from arrests with asystole. Corrupted ECG was constructed by adding a scaled CC artifact to a VF with strength determined by signal-to-interference ratio (SIR). Using empirical mode decomposition, the fluctuation and frequency of CC were retrieved. VF was detected by subtracting the CC artifacts, modeled by least mean-square (LMS) algorithm, from ECG signals. Amplitude spectrum analysis (AMSA) was used to quantify the component with frequency range 0.7[[Unable to Display Character: ‐]]30 Hz. Result: One-hundred and fifty VF (36 successful and 114 unsuccessful defibrillation, p<0.01) were included. After removing CC artifacts with the algorithm, successful vs unsuccessful cases remained distinguished (p<0.05) in SIR ranged 0 to -9dB. The detected and the original VFs had similar AMSA with correlation of 0.98, 0.95, 0.93 and 0.86 when SIR were 0, -3, -6 and -9dB (p<0.001). The mean of difference between the detected VFs and the original VFs were 0.19, 0.56, 1.36 and 3.33 for SIR of 0, -3, -6 and -9dB, respectively, in Bland-Altman plot. Conclusions: The proposed adaptive algorithm using LMS could model CPR artifact without external signals, and effectively detect VF with continuing CC.

Improved Spectrogram Analysis for ECG Signal in Emergency Medical Applications

This paper presents the spectrogram effect of biomedical signal, especially for ECG. Simulation module developed for the spectrogram implementation. Spectrogram based on ECG signal and power spectral density together with off-line evaluation has been observed. ECG contains very important clinical information about the cardiac activities of heart. The features of small variations in ECG signal with time-varying morphological characteristics needs to be extracted by signal processing method because there are not visible of graphical ECG signal. Small variations of simulated normal and noise corrupted ECG signal have been extracted using spectrogram. The spectrogram found to be more precise over conventional FFT in finding the small abnormalities in ECG signal. These form time-frequency representations for processing time-varying signals. By using the presented method, it is ensure that high resolution time-varying spectrum estimation with no lag error can be produced. Other benefits of the method are the straightforward procedure for evaluating the statistics of the spectrum estimation.