An Efficient Kalman Noise Canceller for Cardiac Signal Analysis in Modern Telecardiology Systems

Sulthana, A.K. Thasleem; Rahman, Md. Zıa Ur; Mirza, Shafi Shahsavar

doi:10.1109/access.2018.2848201

Cited by 28 publications

(11 citation statements)

References 33 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…For the NSTDB, because the EM noise widely distributed in the NSTDB distributes in about 1-10Hz [36], the SNR of the ECG with the EM noise in the high-frequency band is higher than that in the low-frequency band. Losing the highfrequency information, the QRS detection algorithms with the conventional bandpass filters (such as [15], [16]) or the wavelet transforms similar to bandpass filters (such as [10]) TABLE 6.…”

Section: B Theoretical Analysis For the Results In Different Databasesmentioning

confidence: 99%

Robust QRS Detection Using High-Resolution Wavelet Packet Decomposition and Time-Attention Convolutional Neural Network

Jia

et al. 2020

IEEE Access

View full text Add to dashboard Cite

QRS detection is a crucial step in analyzing the electrocardiogram (ECG). For ECG collected by wearable devices, a robust QRS detection algorithm that yields high accuracy in spite of abnormal QRS morphologies and severe noise is needed. In this paper, we propose a QRS detection method based on high-resolution wavelet packet decomposition (HR-WPD) and convolutional neural network (CNN). Firstly, we design the HR-WPD that decomposes the ECG into multiple signals with different frequency bands to provide detailed QRS features. Secondly, all the decomposed signals are forwarded to a CNN for comprehensive morphology analysis and QRS prediction. To further improve the robustness, a time-attention module acting on the input signals is added to the CNN. Finally, a variable threshold is imposed to locate the QRS. The proposed method is validated by using two noisy databases (i.e., Telehealth Database (TELEDB) and MIT-BIH Noise Stress Test Database (NSTDB)) and one database with multiple ECG morphologies (i.e., MIT-BIH Arrhythmia Database (ARRDB)). The experiment results show that the proposed method achieves a comparable or even better performance compared with state-of-art methods on the TELEDB (SE 98.99%, P+ 95.57%, ER 5.61%, F1 97.25%), NSTDB (SE 99.25%, P+ 96.31%, ER 4.55%, F1 97.76%) and ARRDB (SE 99.89%, P+ 99.90%, ER 0.21%, F1 99.89%), suggesting that it is highly applicable to the QRS detection for ECG collected by wearable devices. INDEX TERMS Electrocardiogram, convolutional neural network, wavelet packet decomposition, QRS detect.

show abstract

Section: B Theoretical Analysis For the Results In Different Databasesmentioning

confidence: 99%

Robust QRS Detection Using High-Resolution Wavelet Packet Decomposition and Time-Attention Convolutional Neural Network

Jia

et al. 2020

IEEE Access

View full text Add to dashboard Cite

show abstract

“…This variable step size has regularization effects, then its performance is good even for badly excited inputs also. Then by substituting (40) into (12) gives the proposed algorithm (41)…”

Section: Input Noise Variance Estimation Methodsmentioning

confidence: 99%

“…Input variance is estimated if we want to use in BC NLMS algorithm, because it is not practically available. Then practically, BC NLMS is calculated as (12) Where is input noise, variance estimate at particular iteration of t.…”

Section: Normalized Lms Algorithm and Bias Compensated Nlms Algorithmmentioning

confidence: 99%

Artifact Elimination in Thoracic electrical bioimpedance signals using new Normalized Adaptive Filters

2020

IJETER

View full text Add to dashboard Cite

Thoracic electrical bioimpedance (TEB) used for facilitating stroke volume from sudden cardiac arrest signals. It is a non -invasive method used monitoring cardiac outputs, measuring stroke volumes and to observe changes in hemodynamic parameters of volume of blood. While measuring volume of blood, TEB signal is contaminated with physiological and non-physiological signal artifacts. For avoiding these artifacts in this paper proposed an adaptive filter method for enhancing TEB Signals. Least Mean Square (LMS) algorithm is a basic adaptive method, but it is non stationary in nature and it has low convergence rate problems. Hence, Bias compensated Normalized Least Mean Square (BC NLMS) algorithm is proposed, then it check initially for stability in terms of mean deviation analysis and mean square deviation analysis. Depending in this analysis, noisy input variance estimation and variable step size are taken into consideration for better performance in terms of reduced steady error rate, good stability and convergence improvement. In this paper we present various adaptive noise cancellers (ANCs) for the elimination of artifacts from TEB signals. Also, in simulation results, artifacts are eliminated from noisy input signal and it performs well when compared to exiting methods.

show abstract

“…Nowadays, the low-cost biomedical ECG holter or wearable devices are popularly used to monitor the ECG at home, or clinic [3]. However, due to the daily life activity, the ECG waves of these devices are severely contaminated by artifacts such as baseline wander (BW), motion, muscle, and power-line, etc [4]. The different types of methods such as digital filtering, i.e., finite and infinite impulse response (FIR-IIR) filters [5], adaptive filters [6], [7], transform domain filtering such as Fast Fourier Transform (FFT) [8] and Discrete Wavelet Transform (DWT) [9], source decomposition-based filtering such as empirical mode decomposition (EMD) [10] are popularly used by the researchers to remove artifacts from ECG.…”

Section: Introductionmentioning

confidence: 99%

Two Stage Step-Size Scaler Adaptive Filter Design for ECG Denoising

Prajapati

Darji

2021

2021 IEEE International Symposium on Circuits and Systems (ISCAS)

View full text Add to dashboard Cite

In recent years, various healthcare wearable devices have increased to monitor electrocardiogram (ECG) continuously, but due to everyday life activities, artifacts are present in it, which may increases the false prediction of R-peak. This work presents the two-stage active noise cancellation (ANC) stepsize scaler adaptive filter to remove high and low frequencies artifacts from the ECG signal. Moreover, the convergence rate and mean square error (MSE) minimization analysis of step-size scaler adaptive filter in different simulated noise environments is presented and compared with the least mean square (LMS) adaptive filter variants. The analysis shows that the step-size scaler algorithm gives a 67.34% fast convergence rate compared to the normalized LMS (NLMS) algorithm.

show abstract

An Efficient Kalman Noise Canceller for Cardiac Signal Analysis in Modern Telecardiology Systems

Cited by 28 publications

References 33 publications

Robust QRS Detection Using High-Resolution Wavelet Packet Decomposition and Time-Attention Convolutional Neural Network

Robust QRS Detection Using High-Resolution Wavelet Packet Decomposition and Time-Attention Convolutional Neural Network

Artifact Elimination in Thoracic electrical bioimpedance signals using new Normalized Adaptive Filters

Two Stage Step-Size Scaler Adaptive Filter Design for ECG Denoising

Contact Info

Product

Resources

About