ECG signal denoising using higher order statistics in Wavelet subbands

Sharma, L. N.; Dandapat, Samarendra; Mahanta, Anil

doi:10.1016/j.bspc.2010.03.003

Cited by 100 publications

(50 citation statements)

References 18 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…Conventional wavelet schemes are characterized by different parameters that allow the algorithms to be customized for different mixtures of signal and noise sources. The type of the mother wavelet, shrinkage rule, hard versus soft thresholding, noise level rescaling approach and number of decomposition levels are among the different parameters of common wavelet algorithms [20,21,23]. Wavelet transform produces a timefrequency decomposition of the signal under analysis, which separates individual signal components more effectively than the traditional Fourier analysis.…”

Section: Benchmark Methodsmentioning

confidence: 99%

“…Singular value decomposition (SVD) [12][13][14] has also been applied in order to reduce noise in biomedical signals. One of the common approaches is the adaptive filtering (AF) architecture which has been used for interference cancellation of EEG [15][16][17][18][19] and wavelet [20][21][22][23]. In this context, principal component analysis (PCA) [24][25][26][27] and independent component analysis (ICA) [23,[28][29][30][31][32][33][34] have become popular for analysing biomedical data (e.g., EEG and EMG).…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Noise removal in electroencephalogram signals using an artificial neural network based on the simultaneous perturbation method

Mateo

Torres

García

et al. 2015

Neural Comput & Applic

View full text Add to dashboard Cite

Electroencephalogram (EEG) recordings often experience interference by different kinds of noise, including white, muscle and baseline, severely limiting its utility. Artificial neural networks (ANNs) are effective and powerful tools for removing interference from EEGs. Several methods have been developed, but ANNs appear to be the most effective for reducing muscle and baseline contamination, especially when the contamination is greater in amplitude than the brain signal. An ANN as a filter for EEG recordings is proposed in this paper, developing a novel framework for investigating and comparing the relative performance of an ANN incorporating real EEG recordings. This method is based on a growing ANN that optimized the number of nodes in the hidden layer and the coefficient matrices, which are optimized by the simultaneous perturbation method. The ANN improves the results obtained with the conventional EEG filtering techniques: wavelet, singular value decomposition, principal component analysis, adaptive filtering and independent components analysis. The system has been evaluated within a wide range of EEG signals. The present study introduces a new method of reducing all EEG interference signals in one step with low EEG distortion and high noise reduction.

show abstract

Section: Benchmark Methodsmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Noise removal in electroencephalogram signals using an artificial neural network based on the simultaneous perturbation method

Mateo

Torres

García

et al. 2015

Neural Comput & Applic

View full text Add to dashboard Cite

show abstract

“…Estas variaciones crean interferencias de baja frecuencia en el rango de 0 Hz a 0,5 Hz. Estas interferencias en la señal deben ser reducidas para no modificar el resultado de procesos posteriores (Sharma et al, 2010;Kabir and Shahnaz, 2012). En este caso, la primera etapa de procesado de la señal ECG consiste en un filtro de respuesta impulsional infinita (IIR) de orden 8 con una respuesta Butterworth (Kaur et al, 2011;Ravindra Pratap Narwaria and Singhal, 2011).…”

Section: Reducción De Las Oscilaciones De La Línea Baseunclassified

Detección de Fibrilación Ventricular Mediante Tiempo-Frecuencia y Clasificador KNN sin Extracción de Parámetros

Mjahad

Muñoz

Mompean

et al. 2017

Rev. iberoam. autom. inform. ind.

View full text Add to dashboard Cite

ResumenEste trabajo propone la detección de FV y su discriminación de TV y otros ritmos cardiacos basándose en la representación tiempo-frecuencia del ECG y su conversión en imágen como entrada a un clasificador de vecinos más cercanos (KNN) sin necesidad de extracción de parámetros adicionales. Tres variantes de datos de entrada al clasificador son evaluados. Los resultados clasifican la señal en cuatro clases diferentes: 'Normal' para latidos con ritmo sinusal, 'FV' para fibrilación ventricular, 'TV' para taquicardia ventricular y 'Otros' para el resto de ritmos. Los resultados para detección de FV mostraron 88,27 % de sensibilidad y 98,22 % de especificidad para la entrada de imagen equivalente reducida que es la más rápida computacionalmente a pesar de obtener resultados de clasificación ligeramente inferiores a las representaciones no reducidas. En el caso de TV, se alcanzó un 88,31 % de sensibilidad y 98,80 % de especificidad, un 98,14 % de sensibilidad y 96,82 % de especificidad para ritmo sinusal normal y 96,91 % de sensibilidad con 99,06 % de especificidad para la clase 'Otros'. Finalmente, se realiza una comparación con otros algoritmos.Palabras Clave: Sistemas biomédicos, Señales Electrocardiográficas, Representación tiempo-frecuencia, Señales no estacionarias, Análisis de imágenes, Clasificación Ventricular Fibrillation detection using time-frequency and the KNN classifier without parameter extraction. AbstractThis work describes new techniques to improve VF detection and its separation from Ventricular Tachycarida (VT) and other rhythms. It is based on time-frequency representation of the ECG and its use as input in an automatic classifier (K-nearest neighbours -KNN) without any further signal parameter extraction or additional characteristics. For comparison purposes, three time-frequency variants are analysed: pseudo Wigner-Ville representation (RTF), grey-scale image obtained from RTF (IRTF), and reduced image from IRTF (reduced IRTF). Four types of rhythms (classes) are defined: 'Normal' for sinus rhythm, 'VT' for ventricular tachycardia, 'VF' for ventricular fibrillation and 'Others' for the rest of rhythms. Classification results for VF detection in case of reduced IRTF are 88.27 % sensitivity and 98.22 % specificity. In case of VT, 88.31 % sensitivity and 98.80 % specificity is obtained, 98.14 % sensitivity and 96.82 % specificity for normal rhythms, and 96.91 % sensitivity and 99.06 % specificity for other rhythms. Finally, results are compared with other authors.

show abstract

“…Fazlul Haque et al (2009) suggested a detection of small variations of ECG features using wavelet. Sharma et al (2010) developed an ECG signal denoising using higher order statistics in wavelet subbands.…”

Section: Introductionmentioning

confidence: 99%

Denoising of weak ECG signals by using wavelet analysis and fuzzy thresholding

Üstündağ

Gökbulut

Şengür

et al. 2012

Netw Model Anal Health Inform Bioinforma

View full text Add to dashboard Cite

The electrocardiogram (ECG) is a biological signal that contains important information about the cardiac activities of heart. ECG signal plays a very important role in the diagnosis and analysis of heart diseases. ECG signal is corrupted by various types of noise such as electrode movement, strong electromagnetic effect and muscle noise. Noisy ECG signal has been extracted using signal processing. This paper presents a weak ECG signal denoising method based on fuzzy thresholding and wavelet packet analysis. Firstly, the weak ECG signal is decomposed into various levels by wavelet packet transform. Then, the threshold value is determined using the fuzzy s-function. The reconstruction of the ECG signal from the retained coefficients is achieved by using inverse wavelet packet transform. We carried out several experiments to show the effectiveness of the proposed method and compared the results with the traditional wavelet packet soft and hard thresholding methods for weak signal denoising. The results are satisfactory according to calculated the correlation coefficient.

show abstract

ECG signal denoising using higher order statistics in Wavelet subbands

Cited by 100 publications

References 18 publications

Noise removal in electroencephalogram signals using an artificial neural network based on the simultaneous perturbation method

Noise removal in electroencephalogram signals using an artificial neural network based on the simultaneous perturbation method

Detección de Fibrilación Ventricular Mediante Tiempo-Frecuencia y Clasificador KNN sin Extracción de Parámetros

Denoising of weak ECG signals by using wavelet analysis and fuzzy thresholding

Contact Info

Product

Resources

About