The Importance of Gender Specification for Detection of Driver Fatigue using a Single EEG Channel

Mohammad Shahbakhti, Matin Beiramvand, Erfan Nasiri, Wei Chen, Jordi Solé-Casals, Michal Wierzchon, Anna Broniec-Wójcik, Piotr Augustyniak, Vaidotas Marozas

Research output: Chapter in Book/Report/Conference proceedingConference contributionScientificpeer-review


Although detection of the driver fatigue using a single electroencephalography (EEG) channel has been addressed in literature, the gender differentiation for applicability of the model has not been investigated heretofore. Motivated accordingly, we address the detection of driver fatigue based the gender-segregated datasets, where each of them contains 8 subjects. After splitting the EEG signal into its sub-bands (delta, theta, alpha, beta, and gamma) using discrete wavelet transform, the log energy entropy of each band is computed to form the feature vector. Afterwards, the feature vector is randomly split into 50% for training and 50% for the unseen testing, and fed to a support vector machine model. When comparing the classification results of fatigue driving detection between the gender segregated and non-gender segregated datasets, the former achieved the accuracy 78% and 77% for male and female subjects, respectively, than the latter (71%). The obtained results show the importance of gender-specification for the driver fatigue detection.
Original languageEnglish
Title of host publicationBMEiCON 2022 - 14th Biomedical Engineering International Conference
Number of pages3
ISBN (Electronic)978-1-6654-8903-4
Publication statusPublished - 2023
Publication typeA4 Article in conference proceedings
EventBiomedical Engineering International Conference - Songkhla, Thailand
Duration: 10 Nov 202213 Nov 2022


ConferenceBiomedical Engineering International Conference

Publication forum classification

  • Publication forum level 1


Dive into the research topics of 'The Importance of Gender Specification for Detection of Driver Fatigue using a Single EEG Channel'. Together they form a unique fingerprint.

Cite this