Deep Convolutional Neural Networks for mental load classification based on EEG data. (April 2018)

• Record Type: Journal Article
• Authors: Jiao, Zhicheng; Gao, Xinbo; Wang, Ying; Li, Jie; Xu, Haojun
• Is Part Of: Pattern recognition. Volume 76(2018:Apr.)
• Page Start: 582
• Page End: 595
• Publication Date: 2018-04
• Subjects: Deep learning -- Mental load classification -- CNNs -- EEG; Pattern perception -- Periodicals; Perception des structures -- Périodiques; Patroonherkenning; 006.4
• DOI: 10.1016/j.patcog.2017.12.002
• Languages: English
• ISSNs: 0031-3203

Abstract

Highlights: Both single-channel and multi-channel CNN models are developed to obtain representation from spatial and temporal information of EEG data. A point-wise gated Boltzmann machines component is introduced to our models to improve performance of our CNN models. Both our independent and fused models achieve better performance on mental load classification task, and our models contain much less parameters which result in higher efficiency. Abstract: Electroencephalograph (EEG), the representation of the brain's electrical activity, is a widely used measure of brain activities such as working memory during cognitive tasks. Varying in complexity of cognitive tasks, mental load results in different EEG recordings. Classification of mental load is one of core issues in studies on working memory. Various machine learning methods have been introduced into this area, achieving competitive performance. Inspired by the recent breakthrough via deep recurrent convolutional neural networks (CNNs) on classifying mental load, we propose improved CNNs methods for this task. Specifically, our frameworks contain both single-model and double-model methods. With the help of our models, spatial, spectral, and temporal information of EEG data is taken into consideration. Meanwhile, a novel fusion strategy for utilizing different networks is introduced in this work. The proposed methods have been compared with state-of-the-art ones on the same EEG database. The comparison results show that both our single-model method and double-model method can achieve comparable or even better performance than the well-performed deep recurrent CNNs. Furthermore, our proposed CNNs models contain less parameters than state-of-the-art ones, making it be more competitive in further practical application.