Feature extraction of motor imagery EEG signals based on wavelet packet decomposition

Hu, Dingyin; Li, Wei; Chen, Xi

doi:10.1109/iccme.2011.5876829

Cited by 38 publications

(22 citation statements)

References 15 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…Este valor fue el número de características más relevantes en nuestro algoritmo y nos permite reducir a la tercera parte el vector de características inicial, lo que puede traducirse en menor costo computacional durante la clasificación del conjunto de datos, pues al ejecutar los dos algoritmos encontramos que el tiempo de ejecución se reduce en un 14 %. Un análisis de las características más relevantes seleccionadas a partir del algoritmo WP-FE, detalla que los coeficientes Wavelet extraídos del nodo [5,3] presentan la mayor concentración de características, asociadas a los ritmos beta (β), como lo muestra la Tabla III. Es de verse también que las funciones aportadas por la potencia espectral y la varianza de los coeficientes entregan el mayor contenido de información en el vector característico final.…”

Section: Resultados Y Discusionesunclassified

“…Nodos WP Rango de Frecuencia (Hz) [5,2] 7.8125 -15.6250 [5,3] 15.6250 -23.4375 [5,4] 23.4375 -31.2500…”

Section: Tabla I Rango De Frecuencias De Los Nodos Wp Seleccionadosmentioning

confidence: 99%

“…Se han empleado diversas técnicas de extracción de características para tal fin, principalmente, los modelos autorregresivos [2] y [3], patrones espaciales comunes [4], y la Transformada Wavelet [5] y [6]. En este trabajo se presenta un método efectivo para la caracterización de las señales de electroencefalografía en tareas de imaginación motora, usando la Transformada Wavelet Packets, que ha demostrado ser promisorio en el tratamiento de señales EEG [7].…”

Section: Introductionunclassified

See 2 more Smart Citations

Caracterización de Señales EEG mediante Wavelet Packet y Entropía Difusa para Tareas de Imaginación Motora

Medina¹,

López²

2017

Ing.

View full text Add to dashboard Cite

Context: Clinical rhythm analysis on advanced signal processing methods is very important in medical areas such as brain disorder diagnostic, epilepsy, sleep analysis, anesthesia analysis, and more recently in brain-computer interfaces (BCI).Method: Wavelet transform package is used on this work to extract brain rhythms of electroencephalographic signals (EEG) related to motor imagination tasks. We used the Competition BCI 2008 database for this characterization. Using statistical functions we obtained features that characterizes brain rhythms, which are discriminated using different classifiers; they were evaluated using a 10-fold cross validation criteria.Results: The classification accuracy achieved 81.11 % on average, with a degree of agreement of 61 %, indicating a "suitableçoncordance, as it has been reported in the literature. An analysis of relevance showed the concentration of characteristics provided in the nodes as a result of Wavelet decomposition, as well as the characteristics that more information content contribute to improve the separability decision region for the classification task. Conclusions:The proposed method can be used as a reference to support future studies focusing on characterizing EEG signals oriented to the imagination of left and right hand movement, considering that our results proved to compared favourably to those reported in the literature

show abstract

Section: Resultados Y Discusionesunclassified

“…Nodos WP Rango de Frecuencia (Hz) [5,2] 7.8125 -15.6250 [5,3] 15.6250 -23.4375 [5,4] 23.4375 -31.2500…”

Section: Tabla I Rango De Frecuencias De Los Nodos Wp Seleccionadosmentioning

confidence: 99%

Section: Introductionunclassified

See 1 more Smart Citation

Caracterización de Señales EEG mediante Wavelet Packet y Entropía Difusa para Tareas de Imaginación Motora

Medina¹,

López²

2017

Ing.

View full text Add to dashboard Cite

show abstract

“…As can be seen from Table 1, the classification result of the db4 basis function is significantly better than others; therefore, db4 is selected as the best wavelet basis function for the following research. In fact, among different types of basis functions, the Daubechies family is known for its orthogonality property and efficient filter implementation [27]. db4 was found to be the most It can be seen from Figure 4 that the average instantaneous power spectra of electrodes C3 and C4 show distinct ERS/ERD phenomena from 3.5 s to 6.5 s under different hand movement imagination tasks.…”

Section: Selection Of the Optimal Wavelet Packet Basis Functionmentioning

confidence: 99%

“…As can be seen from Table 1, the classification result of the db4 basis function is significantly better than others; therefore, db4 is selected as the best wavelet basis function for the following research. In fact, among different types of basis functions, the Daubechies family is known for its orthogonality property and efficient filter implementation [27]. db4 was found to be the most appropriate basis for analysis of EEG data since the lower order wavelets of the family was too coarse to represent EEG spike properly, and the higher one cannot represent the spiky form of the EEG signal [28].…”

Section: Selection Of the Optimal Wavelet Packet Basis Functionmentioning

confidence: 99%

Adaptive Feature Extraction of Motor Imagery EEG with Optimal Wavelet Packets and SE-Isomap

Zhu

Liu

et al. 2017

Applied Sciences

View full text Add to dashboard Cite

Abstract:Motor imagery EEG (MI-EEG), which reflects one's active movement intention, has attracted increasing attention in rehabilitation therapy, and accurate and fast feature extraction is the key problem to successful applications. Based on wavelet packet decomposition (WPD) and SE-isomap, an adaptive feature extraction method is proposed in this paper. The MI-EEG is preprocessed to determine a more effective time interval through average power spectrum analysis. WPD is then applied to the selected segment of MI-EEG, and the subject-based optimal wavelet packets (OWPs) with top mean variance difference are obtained autonomously. The OWP coefficients are further used to calculate the time-frequency features statistically and acquire the nonlinear manifold structure features, as well as the explicit nonlinear mapping, through SE-isomap. The hybrid features are obtained in a serial fusion way and evaluated by a k-nearest neighbor (KNN) classifier. The extensive experiments are conducted on a publicly available dataset, and the experiment results of 10-fold cross-validation show that the proposed method yields relatively higher classification accuracy and computation efficiency simultaneously compared with the commonly-used linear and nonlinear approaches.

show abstract

Impact of Noise Elimination Methods on Classification Performance in Motor Imagery EEG

Özkahraman,

Ölmez,

Dokur

2024

Lecture Notes in Networks and Systems

View full text Add to dashboard Cite

Feature extraction of motor imagery EEG signals based on wavelet packet decomposition

Cited by 38 publications

References 15 publications

Caracterización de Señales EEG mediante Wavelet Packet y Entropía Difusa para Tareas de Imaginación Motora

Caracterización de Señales EEG mediante Wavelet Packet y Entropía Difusa para Tareas de Imaginación Motora

Adaptive Feature Extraction of Motor Imagery EEG with Optimal Wavelet Packets and SE-Isomap

Impact of Noise Elimination Methods on Classification Performance in Motor Imagery EEG

Contact Info

Product

Resources

About