Learning Temporal Information for Brain-Computer Interface Using Convolutional Neural Networks

By Siavash Sakhavi, Cuntai Guan, Shuicheng Yan. IEEE Transactions on Neural Networks and Learning Systems, 2018

545 citations recorded by the source corpus.

Deep learning (DL) methods and architectures have been the state-of-the-art classification algorithms for computer vision and natural language processing problems. However, the successful application of these methods in motor imagery (MI) brain–computer interfaces (BCIs), in order to boost classification performance, is still limited. In this paper, we propose a classification framework for MI data by introducing a new temporal representation of the data and also utilizing a convolutional neural network (CNN) architecture for classification. The new representation is generated from modifying the filter-bank common spatial patterns method, and the CNN is designed and optimized accordingly for the representation. Our framework outperforms the best classification method in the literature on the BCI competition IV-2a 4-class MI data set by 7% increase in average subject accuracy. Furthermore, by studying the convolutional weights of the trained networks, we gain an insight into the temporal characteristics of EEG.

Subfields: Cognitive Neuroscience, Electrical and Electronic Engineering, Signal Processing

View Paper: https://www.semanticscholar.org/paper/656964e969bc2b99a8e18c8fc9372bab71e12b67

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Cognitive Neuroscience

In Electrical and Electronic Engineering

In Signal Processing

Citations in the corpus

EEG Conformer: Convolutional Transformer for EEG Decoding and Visualization (2022, methodology): “Sakhavi et al. [28] used CNN to learn temporal information from the ﬁlter bank CSP features and select architecture parameters for each subject.”
EEG-Based Spatio–Temporal Convolutional Neural Network for Driver Fatigue Evaluation (2019, unknown): “outperformed the existing results on the motor imagery data set [31].”
CNN-RNN Based Intelligent Recommendation for Online Medical Pre-Diagnosis Support (2020, methodology): “Sakhavi et al. [21] designed a classiﬁcation framework for motor imagery, based on a CNN architecture, which was optimized in accordance with a temporal representation with spatial patterns to pursue a better classiﬁcation result.”
EEG-Based Brain-Computer Interfaces (BCIs): A Survey of Recent Studies on Signal Sensing Technologies and Computational Intelligence Approaches and Their Applications (2020, methodology): “The MI classification framework proposed by Sakhavi, Guan and Yan [109] presents a new data temporal representation gener-”
Federated Transfer Learning for EEG Signal Classification (2020, methodology): “Furthermore, deep learning methods, especially the convolutional neural network (CNN) architectures, have been explored to boost classification performance [18], [19], [20], [21].”
Deep learning techniques for classification of electroencephalogram (EEG) motor imagery (MI) signals: a review (2021, unknown): “2018, [72] ALL (22) 4–40 N/A EF: CSP CNN 3 CONV 1 FC 4 OUT ReLU: conv Smax: L-FC”
Subject-Independent Brain–Computer Interfaces Based on Deep Convolutional Neural Networks (2019, background): “Due to the lack of a large public MI database, the previous studies using DL could work only in subject-dependent environments [48]–[51].”
TSception: Capturing Temporal Dynamics and Spatial Asymmetry From EEG for Emotion Recognition (2021, methodology): “…fusion to detect emotions from EEG. Recently, deep learning-based methods have shown promising results in the BCI domain, such as motor imagery clas-siﬁcation [15], [16], [17], [18], [19], emotion recognition [20], [21], [22], [23], [24], [25], and mental-task classiﬁcation [26], [27], [28].”
Artificial Intelligence for the Metaverse: A Survey (2022, background): “, linear discriminant analysis and CNN [133]) but also obtained the practicality with differ-”