Implementation of wavelet packet transform and non linear analysis for emotion classification in stroke patient using brain signals

Siao Zheng Bong, Khairunizam Wan, M. Murugappan, Norlinah Mohamed Ibrahim, Yuvaraj Rajamanickam, Khairiyah Mohamad

Research output: Contribution to journalArticle

14 Citations (Scopus)

Abstract

Emotion perception in stroke patients is affected since there is abnormality in the brain. Here, researchers focused on the impact of left brain damage and right brain damage towards emotion recognition. Due to the impaired emotion recognition, it is a challenge for stroke patients to express themselves in daily communication. Hence, it is inspiring to see the possibility to predict patient's emotional state so as to prevent recurrent stroke. In this work, electroencephalograph (EEG) of 19 left brain damage patients (LBD), 19 right brain damage patients (RBD) and 19 normal control (NC) are collected as database. During data collection, six emotions (sad, disgust, fear, anger, happy and surprise) are induced by using audio visual stimuli. After normalization, EEG signals are filtered by using Butterworth 6th order band-pass filter at the cut-off frequencies of 0.5 Hz and 49 Hz. Then, wavelet packet transform (WPT) technique is implemented to localize five frequency bands: alpha (8 Hz–13 Hz), beta (13 Hz–30 Hz), gamma (30 Hz–49 Hz), alpha-to-gamma (8 Hz–49 Hz), beta-to-gamma (13 Hz–49 Hz). In WPT, four wavelet families are chosen: daubechies 4 (db4), daubechies 6 (db6), coiflet 5 (coif5) and symmlet 8 (sym8). Hurst exponents are extracted from each band and wavelet family and are classified by using K-nearest Neighbour (KNN) and Probabilistic Neural Network (PNN). Two classifications are done: comparison between three groups and comparison between six emotions. The results showed that all the H values are anti-correlated (0 < H < 0.5). From classification, the best frequency band is beta band, where sad emotion recorded the accuracy of 82.32% for LBD group. Meanwhile, both sad and fear emotion recorded 0.89 sensitivity score in LBD and RBD respectively. Due to its overall poor performance, RBD is found to have greater impairment in emotion recognition.

Original languageEnglish
Pages (from-to)102-112
Number of pages11
JournalBiomedical Signal Processing and Control
Volume36
DOIs
Publication statusPublished - 1 Jul 2017
Externally publishedYes

Fingerprint

Wavelet Analysis
Nonlinear analysis
Brain
Emotions
Stroke
Patient Rights
Frequency bands
Fear
Anger
Cutoff frequency
Bandpass filters
Communication
Research Personnel
Databases
Neural networks

Keywords

  • Electroencephalogram (EEG)
  • Emotion recognition
  • K-Nearest neighbour (KNN)
  • Probabilistic neural network (PNN)
  • Stroke
  • Wavelet packet transform (WPT)

ASJC Scopus subject areas

  • Signal Processing
  • Health Informatics

Cite this

Implementation of wavelet packet transform and non linear analysis for emotion classification in stroke patient using brain signals. / Bong, Siao Zheng; Wan, Khairunizam; Murugappan, M.; Mohamed Ibrahim, Norlinah; Rajamanickam, Yuvaraj; Mohamad, Khairiyah.

In: Biomedical Signal Processing and Control, Vol. 36, 01.07.2017, p. 102-112.

Research output: Contribution to journalArticle

Bong, Siao Zheng ; Wan, Khairunizam ; Murugappan, M. ; Mohamed Ibrahim, Norlinah ; Rajamanickam, Yuvaraj ; Mohamad, Khairiyah. / Implementation of wavelet packet transform and non linear analysis for emotion classification in stroke patient using brain signals. In: Biomedical Signal Processing and Control. 2017 ; Vol. 36. pp. 102-112.
@article{0696314e4f8046569baa0e77e17f866b,
title = "Implementation of wavelet packet transform and non linear analysis for emotion classification in stroke patient using brain signals",
abstract = "Emotion perception in stroke patients is affected since there is abnormality in the brain. Here, researchers focused on the impact of left brain damage and right brain damage towards emotion recognition. Due to the impaired emotion recognition, it is a challenge for stroke patients to express themselves in daily communication. Hence, it is inspiring to see the possibility to predict patient's emotional state so as to prevent recurrent stroke. In this work, electroencephalograph (EEG) of 19 left brain damage patients (LBD), 19 right brain damage patients (RBD) and 19 normal control (NC) are collected as database. During data collection, six emotions (sad, disgust, fear, anger, happy and surprise) are induced by using audio visual stimuli. After normalization, EEG signals are filtered by using Butterworth 6th order band-pass filter at the cut-off frequencies of 0.5 Hz and 49 Hz. Then, wavelet packet transform (WPT) technique is implemented to localize five frequency bands: alpha (8 Hz–13 Hz), beta (13 Hz–30 Hz), gamma (30 Hz–49 Hz), alpha-to-gamma (8 Hz–49 Hz), beta-to-gamma (13 Hz–49 Hz). In WPT, four wavelet families are chosen: daubechies 4 (db4), daubechies 6 (db6), coiflet 5 (coif5) and symmlet 8 (sym8). Hurst exponents are extracted from each band and wavelet family and are classified by using K-nearest Neighbour (KNN) and Probabilistic Neural Network (PNN). Two classifications are done: comparison between three groups and comparison between six emotions. The results showed that all the H values are anti-correlated (0 < H < 0.5). From classification, the best frequency band is beta band, where sad emotion recorded the accuracy of 82.32{\%} for LBD group. Meanwhile, both sad and fear emotion recorded 0.89 sensitivity score in LBD and RBD respectively. Due to its overall poor performance, RBD is found to have greater impairment in emotion recognition.",
keywords = "Electroencephalogram (EEG), Emotion recognition, K-Nearest neighbour (KNN), Probabilistic neural network (PNN), Stroke, Wavelet packet transform (WPT)",
author = "Bong, {Siao Zheng} and Khairunizam Wan and M. Murugappan and {Mohamed Ibrahim}, Norlinah and Yuvaraj Rajamanickam and Khairiyah Mohamad",
year = "2017",
month = "7",
day = "1",
doi = "10.1016/j.bspc.2017.03.016",
language = "English",
volume = "36",
pages = "102--112",
journal = "Biomedical Signal Processing and Control",
issn = "1746-8094",
publisher = "Elsevier BV",

}

TY - JOUR

T1 - Implementation of wavelet packet transform and non linear analysis for emotion classification in stroke patient using brain signals

AU - Bong, Siao Zheng

AU - Wan, Khairunizam

AU - Murugappan, M.

AU - Mohamed Ibrahim, Norlinah

AU - Rajamanickam, Yuvaraj

AU - Mohamad, Khairiyah

PY - 2017/7/1

Y1 - 2017/7/1

N2 - Emotion perception in stroke patients is affected since there is abnormality in the brain. Here, researchers focused on the impact of left brain damage and right brain damage towards emotion recognition. Due to the impaired emotion recognition, it is a challenge for stroke patients to express themselves in daily communication. Hence, it is inspiring to see the possibility to predict patient's emotional state so as to prevent recurrent stroke. In this work, electroencephalograph (EEG) of 19 left brain damage patients (LBD), 19 right brain damage patients (RBD) and 19 normal control (NC) are collected as database. During data collection, six emotions (sad, disgust, fear, anger, happy and surprise) are induced by using audio visual stimuli. After normalization, EEG signals are filtered by using Butterworth 6th order band-pass filter at the cut-off frequencies of 0.5 Hz and 49 Hz. Then, wavelet packet transform (WPT) technique is implemented to localize five frequency bands: alpha (8 Hz–13 Hz), beta (13 Hz–30 Hz), gamma (30 Hz–49 Hz), alpha-to-gamma (8 Hz–49 Hz), beta-to-gamma (13 Hz–49 Hz). In WPT, four wavelet families are chosen: daubechies 4 (db4), daubechies 6 (db6), coiflet 5 (coif5) and symmlet 8 (sym8). Hurst exponents are extracted from each band and wavelet family and are classified by using K-nearest Neighbour (KNN) and Probabilistic Neural Network (PNN). Two classifications are done: comparison between three groups and comparison between six emotions. The results showed that all the H values are anti-correlated (0 < H < 0.5). From classification, the best frequency band is beta band, where sad emotion recorded the accuracy of 82.32% for LBD group. Meanwhile, both sad and fear emotion recorded 0.89 sensitivity score in LBD and RBD respectively. Due to its overall poor performance, RBD is found to have greater impairment in emotion recognition.

AB - Emotion perception in stroke patients is affected since there is abnormality in the brain. Here, researchers focused on the impact of left brain damage and right brain damage towards emotion recognition. Due to the impaired emotion recognition, it is a challenge for stroke patients to express themselves in daily communication. Hence, it is inspiring to see the possibility to predict patient's emotional state so as to prevent recurrent stroke. In this work, electroencephalograph (EEG) of 19 left brain damage patients (LBD), 19 right brain damage patients (RBD) and 19 normal control (NC) are collected as database. During data collection, six emotions (sad, disgust, fear, anger, happy and surprise) are induced by using audio visual stimuli. After normalization, EEG signals are filtered by using Butterworth 6th order band-pass filter at the cut-off frequencies of 0.5 Hz and 49 Hz. Then, wavelet packet transform (WPT) technique is implemented to localize five frequency bands: alpha (8 Hz–13 Hz), beta (13 Hz–30 Hz), gamma (30 Hz–49 Hz), alpha-to-gamma (8 Hz–49 Hz), beta-to-gamma (13 Hz–49 Hz). In WPT, four wavelet families are chosen: daubechies 4 (db4), daubechies 6 (db6), coiflet 5 (coif5) and symmlet 8 (sym8). Hurst exponents are extracted from each band and wavelet family and are classified by using K-nearest Neighbour (KNN) and Probabilistic Neural Network (PNN). Two classifications are done: comparison between three groups and comparison between six emotions. The results showed that all the H values are anti-correlated (0 < H < 0.5). From classification, the best frequency band is beta band, where sad emotion recorded the accuracy of 82.32% for LBD group. Meanwhile, both sad and fear emotion recorded 0.89 sensitivity score in LBD and RBD respectively. Due to its overall poor performance, RBD is found to have greater impairment in emotion recognition.

KW - Electroencephalogram (EEG)

KW - Emotion recognition

KW - K-Nearest neighbour (KNN)

KW - Probabilistic neural network (PNN)

KW - Stroke

KW - Wavelet packet transform (WPT)

UR - http://www.scopus.com/inward/record.url?scp=85017268497&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85017268497&partnerID=8YFLogxK

U2 - 10.1016/j.bspc.2017.03.016

DO - 10.1016/j.bspc.2017.03.016

M3 - Article

AN - SCOPUS:85017268497

VL - 36

SP - 102

EP - 112

JO - Biomedical Signal Processing and Control

JF - Biomedical Signal Processing and Control

SN - 1746-8094

ER -