Optimal set of EEG features for emotional state classification and trajectory visualization in Parkinson's disease

R. Yuvaraj, M. Murugappan, Norlinah Mohamed Ibrahim, Kenneth Sundaraj, Mohd I qbal Omar, Khairiyah Mohamad, R. Palaniappan

Research output: Contribution to journalArticle

27 Citations (Scopus)

Abstract

In addition to classic motor signs and symptoms, individuals with Parkinson's disease (PD) are characterized by emotional deficits. Ongoing brain activity can be recorded by electroencephalograph (EEG) to discover the links between emotional states and brain activity. This study utilized machine-learning algorithms to categorize emotional states in PD patients compared with healthy controls (HC) using EEG. Twenty non-demented PD patients and 20 healthy age-, gender-, and education level-matched controls viewed happiness, sadness, fear, anger, surprise, and disgust emotional stimuli while fourteen-channel EEG was being recorded. Multimodal stimulus (combination of audio and visual) was used to evoke the emotions. To classify the EEG-based emotional states and visualize the changes of emotional states over time, this paper compares four kinds of EEG features for emotional state classification and proposes an approach to track the trajectory of emotion changes with manifold learning. From the experimental results using our EEG data set, we found that (a) bispectrum feature is superior to other three kinds of features, namely power spectrum, wavelet packet and nonlinear dynamical analysis; (b) higher frequency bands (alpha, beta and gamma) play a more important role in emotion activities than lower frequency bands (delta and theta) in both groups and; (c) the trajectory of emotion changes can be visualized by reducing subject-independent features with manifold learning. This provides a promising way of implementing visualization of patient's emotional state in real time and leads to a practical system for noninvasive assessment of the emotional impairments associated with neurological disorders.

Original languageEnglish
Pages (from-to)482-495
Number of pages14
JournalInternational Journal of Psychophysiology
Volume94
Issue number3
DOIs
Publication statusPublished - 1 Dec 2014

Fingerprint

Parkinson Disease
Emotions
Learning
Happiness
Anger
Brain
Nervous System Diseases
Signs and Symptoms
Fear
Education

Keywords

  • Electroencephalogram
  • Emotion classification
  • Feature reduction
  • Manifold learning
  • Parkinson's disease

ASJC Scopus subject areas

  • Medicine(all)

Cite this

Optimal set of EEG features for emotional state classification and trajectory visualization in Parkinson's disease. / Yuvaraj, R.; Murugappan, M.; Mohamed Ibrahim, Norlinah; Sundaraj, Kenneth; Omar, Mohd I qbal; Mohamad, Khairiyah; Palaniappan, R.

In: International Journal of Psychophysiology, Vol. 94, No. 3, 01.12.2014, p. 482-495.

Research output: Contribution to journalArticle

Yuvaraj, R. ; Murugappan, M. ; Mohamed Ibrahim, Norlinah ; Sundaraj, Kenneth ; Omar, Mohd I qbal ; Mohamad, Khairiyah ; Palaniappan, R. / Optimal set of EEG features for emotional state classification and trajectory visualization in Parkinson's disease. In: International Journal of Psychophysiology. 2014 ; Vol. 94, No. 3. pp. 482-495.
@article{ba9c7f767c3643b8b31cebbbc79245d9,
title = "Optimal set of EEG features for emotional state classification and trajectory visualization in Parkinson's disease",
abstract = "In addition to classic motor signs and symptoms, individuals with Parkinson's disease (PD) are characterized by emotional deficits. Ongoing brain activity can be recorded by electroencephalograph (EEG) to discover the links between emotional states and brain activity. This study utilized machine-learning algorithms to categorize emotional states in PD patients compared with healthy controls (HC) using EEG. Twenty non-demented PD patients and 20 healthy age-, gender-, and education level-matched controls viewed happiness, sadness, fear, anger, surprise, and disgust emotional stimuli while fourteen-channel EEG was being recorded. Multimodal stimulus (combination of audio and visual) was used to evoke the emotions. To classify the EEG-based emotional states and visualize the changes of emotional states over time, this paper compares four kinds of EEG features for emotional state classification and proposes an approach to track the trajectory of emotion changes with manifold learning. From the experimental results using our EEG data set, we found that (a) bispectrum feature is superior to other three kinds of features, namely power spectrum, wavelet packet and nonlinear dynamical analysis; (b) higher frequency bands (alpha, beta and gamma) play a more important role in emotion activities than lower frequency bands (delta and theta) in both groups and; (c) the trajectory of emotion changes can be visualized by reducing subject-independent features with manifold learning. This provides a promising way of implementing visualization of patient's emotional state in real time and leads to a practical system for noninvasive assessment of the emotional impairments associated with neurological disorders.",
keywords = "Electroencephalogram, Emotion classification, Feature reduction, Manifold learning, Parkinson's disease",
author = "R. Yuvaraj and M. Murugappan and {Mohamed Ibrahim}, Norlinah and Kenneth Sundaraj and Omar, {Mohd I qbal} and Khairiyah Mohamad and R. Palaniappan",
year = "2014",
month = "12",
day = "1",
doi = "10.1016/j.ijpsycho.2014.07.014",
language = "English",
volume = "94",
pages = "482--495",
journal = "International Journal of Psychophysiology",
issn = "0167-8760",
publisher = "Elsevier",
number = "3",

}

TY - JOUR

T1 - Optimal set of EEG features for emotional state classification and trajectory visualization in Parkinson's disease

AU - Yuvaraj, R.

AU - Murugappan, M.

AU - Mohamed Ibrahim, Norlinah

AU - Sundaraj, Kenneth

AU - Omar, Mohd I qbal

AU - Mohamad, Khairiyah

AU - Palaniappan, R.

PY - 2014/12/1

Y1 - 2014/12/1

N2 - In addition to classic motor signs and symptoms, individuals with Parkinson's disease (PD) are characterized by emotional deficits. Ongoing brain activity can be recorded by electroencephalograph (EEG) to discover the links between emotional states and brain activity. This study utilized machine-learning algorithms to categorize emotional states in PD patients compared with healthy controls (HC) using EEG. Twenty non-demented PD patients and 20 healthy age-, gender-, and education level-matched controls viewed happiness, sadness, fear, anger, surprise, and disgust emotional stimuli while fourteen-channel EEG was being recorded. Multimodal stimulus (combination of audio and visual) was used to evoke the emotions. To classify the EEG-based emotional states and visualize the changes of emotional states over time, this paper compares four kinds of EEG features for emotional state classification and proposes an approach to track the trajectory of emotion changes with manifold learning. From the experimental results using our EEG data set, we found that (a) bispectrum feature is superior to other three kinds of features, namely power spectrum, wavelet packet and nonlinear dynamical analysis; (b) higher frequency bands (alpha, beta and gamma) play a more important role in emotion activities than lower frequency bands (delta and theta) in both groups and; (c) the trajectory of emotion changes can be visualized by reducing subject-independent features with manifold learning. This provides a promising way of implementing visualization of patient's emotional state in real time and leads to a practical system for noninvasive assessment of the emotional impairments associated with neurological disorders.

AB - In addition to classic motor signs and symptoms, individuals with Parkinson's disease (PD) are characterized by emotional deficits. Ongoing brain activity can be recorded by electroencephalograph (EEG) to discover the links between emotional states and brain activity. This study utilized machine-learning algorithms to categorize emotional states in PD patients compared with healthy controls (HC) using EEG. Twenty non-demented PD patients and 20 healthy age-, gender-, and education level-matched controls viewed happiness, sadness, fear, anger, surprise, and disgust emotional stimuli while fourteen-channel EEG was being recorded. Multimodal stimulus (combination of audio and visual) was used to evoke the emotions. To classify the EEG-based emotional states and visualize the changes of emotional states over time, this paper compares four kinds of EEG features for emotional state classification and proposes an approach to track the trajectory of emotion changes with manifold learning. From the experimental results using our EEG data set, we found that (a) bispectrum feature is superior to other three kinds of features, namely power spectrum, wavelet packet and nonlinear dynamical analysis; (b) higher frequency bands (alpha, beta and gamma) play a more important role in emotion activities than lower frequency bands (delta and theta) in both groups and; (c) the trajectory of emotion changes can be visualized by reducing subject-independent features with manifold learning. This provides a promising way of implementing visualization of patient's emotional state in real time and leads to a practical system for noninvasive assessment of the emotional impairments associated with neurological disorders.

KW - Electroencephalogram

KW - Emotion classification

KW - Feature reduction

KW - Manifold learning

KW - Parkinson's disease

UR - http://www.scopus.com/inward/record.url?scp=84939435152&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84939435152&partnerID=8YFLogxK

U2 - 10.1016/j.ijpsycho.2014.07.014

DO - 10.1016/j.ijpsycho.2014.07.014

M3 - Article

C2 - 25109433

AN - SCOPUS:84939435152

VL - 94

SP - 482

EP - 495

JO - International Journal of Psychophysiology

JF - International Journal of Psychophysiology

SN - 0167-8760

IS - 3

ER -