Fractional Fourier Transform Aided Computerized Framework for Alcoholism Identification in EEG
Paper
Paper/Presentation Title | Fractional Fourier Transform Aided Computerized Framework for Alcoholism Identification in EEG |
---|---|
Presentation Type | Paper |
Authors | Sadiq, Muhammad Tariq (Author), Akbari, Hesam (Author), Siuly, Siuly (Author), Li, Yan (Author) and Wen, Paul (Author) |
Journal or Proceedings Title | Proceedings of the 11th International Conference on Health Information Science (HIS 2022) |
ERA Conference ID | 73271 |
Journal Citation | 13705, pp. 100-112 |
Number of Pages | 13 |
Year | 2022 |
Place of Publication | Switzerland |
ISBN | 9783031206269 |
9783031206276 | |
Digital Object Identifier (DOI) | https://doi.org/10.1007/978-3-031-20627-6_10 |
Web Address (URL) of Paper | https://link.springer.com/chapter/10.1007/978-3-031-20627-6_10 |
Web Address (URL) of Conference Proceedings | https://link.springer.com/book/10.1007/978-3-031-20627-6 |
Conference/Event | 11th International Conference on Health Information Science (HIS 2022) |
International Conference on Health Information Science (HIS) | |
Event Details | 11th International Conference on Health Information Science (HIS 2022) Parent International Conference on Health Information Science (HIS) Event Date 28 to end of 30 Oct 2022 Event Location Biarritz, France |
Event Details | International Conference on Health Information Science (HIS) HIS |
Abstract | Alcoholism has a detrimental impact on brain functioning. Electroencephalogram (EEG) signals are commonly used by clinicians and researchers to quantify and document alcoholic brain activity. Despite widespread attention in these signals, the non-stationarity of physiological EEG signals has complications in alcoholism applications. Fourier Transform have been used to examine stationary signals in a straightforward manner. Non-stationary signal analysis, on the other hand, is unsatisfactory using such an approach because it cannot show the occurrence time of distinct frequency components. Furthermore, it is critical to capture both time and frequency characteristics. To overcome these aforementioned issues in alcoholism EEG signals, a computer-aided diagnosis (CAD) approach is proposed in this study to distinguish between normal and alcoholic subjects. The dataset is first split into multiple EEG signals, and the multiscale principal component analysis approach is used to remove noises. Second, as a novel and powerful feature extraction method for EEG signals, the Fractional Fourier Transform (FrFT) methodology with different coefficients is used. A generalization of the classical Fourier Transform, the FrFT, may reveal the fluctuating frequencies of non-stationary EEG signals. The t-test method is used to evaluate the FrFT derived coefficients as features. Finally, to categorize normal vs alcoholic signals, relevant features are tested on multiple machine learning classifiers accessible in the WEKA platform using a 10-fold cross-validation technique. The obtained results effectively support the usefulness of FrFT coefficients as features. |
Keywords | Electroencephalography; Computer-aided Diagnosis; Alcoholism ; Fractional Fourier Transform; Non-Stationary; Classification |
ANZSRC Field of Research 2020 | 460299. Artificial intelligence not elsewhere classified |
460304. Computer vision | |
460102. Applications in health | |
Public Notes | Files associated with this item cannot be displayed due to copyright restrictions. |
Series | Lecture Notes in Computer Science |
Byline Affiliations | University of Brighton, United Kingdom |
Islamic Azad University, Iran | |
Victoria University | |
School of Mathematics, Physics and Computing | |
School of Engineering | |
Institution of Origin | University of Southern Queensland |
https://research.usq.edu.au/item/q7qwq/fractional-fourier-transform-aided-computerized-framework-for-alcoholism-identification-in-eeg
79
total views2
total downloads1
views this month0
downloads this month