Autism Spectrum Disorder Diagnostic System Using HOS Bispectrum with EEG Signals

Autor: The-Hanh Pham, Shu Lih Oh, Edward J. Ciaccio, Enas Abdulhay, Joel Koh En Wei, Jahmunah Vicnesh, U. Rajendra Acharya, N. Arunkumar
Rok vydání: 2020
Předmět:
Male
Decision support system
Adolescent
Computer science
Health
Toxicology and Mutagenesis
lcsh:Medicine
autism spectrum disorder
02 engineering and technology
computer-aided brain diagnostic system
Electroencephalography
Article
03 medical and health sciences
Probabilistic neural network
0302 clinical medicine
0202 electrical engineering
electronic engineering
information engineering
medicine
Humans
eeg signals
Child
medicine.diagnostic_test
business.industry
lcsh:R
Public Health
Environmental and Occupational Health
Discriminant Analysis
Signal Processing
Computer-Assisted
classifiers
Pattern recognition
t-test
medicine.disease
Linear discriminant analysis
Autism spectrum disorder
Child
Preschool
nonlinear features
10-fold validation
Autism
Female
020201 artificial intelligence & image processing
Neural Networks
Computer
higher-order spectra bispectrum
Artificial intelligence
locality sensitivity discriminant analysis
business
Bispectrum
Classifier (UML)
030217 neurology & neurosurgery
Zdroj: International Journal of Environmental Research and Public Health, Vol 17, Iss 3, p 971 (2020)
International Journal of Environmental Research and Public Health
Volume 17
Issue 3
ISSN: 1660-4601
DOI: 10.3390/ijerph17030971
Popis: Autistic individuals often have difficulties expressing or controlling emotions and have poor eye contact, among other symptoms. The prevalence of autism is increasing globally, posing a need to address this concern. Current diagnostic systems have particular limitations
hence, some individuals go undiagnosed or the diagnosis is delayed. In this study, an effective autism diagnostic system using electroencephalogram (EEG) signals, which are generated from electrical activity in the brain, was developed and characterized. The pre-processed signals were converted to two-dimensional images using the higher-order spectra (HOS) bispectrum. Nonlinear features were extracted thereafter, and then reduced using locality sensitivity discriminant analysis (LSDA). Significant features were selected from the condensed feature set using Student's t-test, and were then input to different classifiers. The probabilistic neural network (PNN) classifier achieved the highest accuracy of 98.70% with just five features. Ten-fold cross-validation was employed to evaluate the performance of the classifier. It was shown that the developed system can be useful as a decision support tool to assist healthcare professionals in diagnosing autism.
Databáze: OpenAIRE
Externí odkaz: