Citation: Liang, X.; Liu, Y.; Yu, Y.; Liu,
K.; Liu, Y.; Zhou, Z. Convolutional
Neural Network with a Topographic
Representation Module for
EEG-Based Brain—Computer
Interfaces. Brain Sci. 2023, 13, 268.
https://doi.org/10.3390/
brainsci13020268
Academic Editors: Enrico Vezzetti,
Andrea Luigi Guerra, Gabriele
Baronio, Domenico Speranza
and Luca Ulrich
Received: 9 January 2023
Revised: 2 February 2023
Accepted: 3 February 2023
Published: 5 February 2023
Copyright: © 2023 by the authors.
Licensee MDPI, Basel, Switzerland.
This article is an open access article
distributed under the terms and
conditions of the Creative Commons
Attribution (CC BY) license (https://
creativecommons.org/licenses/by/
4.0/).
Article
Convolutional Neural Network with a Topographic Representation
Module for EEG-Based Brain—Computer Interfaces
Xinbin Liang , Yaru Liu, Yang Yu *, Kaixuan Liu, Yadong Liu and Zongtan Zhou
College of Intelligence Science and Technology, National University of Defense Technology,
Changsha 410073, China
* Correspondence: yuyangnudt@hotmail.com
Abstract:
Convolutional neural networks (CNNs) have shown great potential in the field of
brain–computer
interfaces (BCIs) due to their ability to directly process raw electroencephalogram
(EEG) signals without artificial feature extraction. Some CNNs have achieved better classification
accuracy than that of traditional methods. Raw EEG signals are usually represented as a two-
dimensional (2-D) matrix composed of channels and time points, ignoring the spatial topological
information of electrodes. Our goal is to make a CNN that takes raw EEG signals as inputs have the
ability to learn spatial topological features and improve its classification performance while basically
maintaining its original structure. We propose an EEG topographic representation module (TRM).
This module consists of (1) a mapping block from raw EEG signals to a 3-D topographic map and
(2) a convolution block from the topographic map to an output with the same size as the input. Ac-
cording to the size of the convolutional kernel used in the convolution block, we design two types of
TRMs, namely TRM-(5,5) and TRM-(3,3). We embed the two TRM types into three widely used CNNs
(ShallowConvNet, DeepConvNet and EEGNet) and test them on two publicly available datasets (the
Emergency Braking During Simulated Driving Dataset (EBDSDD) and the High Gamma Dataset
(HGD)). Results show that the classification accuracies of all three CNNs are improved on both
datasets after using the TRMs. With TRM-(5,5), the average classification accuracies of DeepConvNet,
EEGNet and ShallowConvNet are improved by 6.54%, 1.72% and 2.07% on the EBDSDD and by
6.05%, 3.02% and 5.14% on the HGD, respectively; with TRM-(3,3), they are improved by 7.76%, 1.71%
and 2.17% on the EBDSDD and by 7.61%, 5.06% and 6.28% on the HGD, respectively. We improve the
classification performance of three CNNs on both datasets through the use of TRMs, indicating that
they have the capability to mine spatial topological EEG information. More importantly, since the
output of a TRM has the same size as the input, CNNs with raw EEG signals as inputs can use this
module without changing their original structures.
Keywords:
convolutional neural network (CNN); electroencephalogram (EEG); topographic
representation; brain–computer interface (BCI); EEG decoding; deep learning
1. Introduction
Brain–computer interfaces (BCIs) enable direct communication between humans and
machines via electroencephalogram (EEG) signals [
1
]. EEG signals contain instinctive
biometric information from the human brain. Through precise EEG decoding, BCIs can
recognize the inner thoughts of users. The EEG applications in ML/DL-based disease,
mental workload and sleep stage prediction have also been widely studied [
2
–
6
]. They are
also based on the classification of EEG signals and are similar to BCIs in implementation and
processing methods. In general, EEG decoding consists of five main stages: data collection,
signal preprocessing, feature extraction, classification and data analysis [
7
]. Although
these stages are essentially the same in a BCI paradigm, signal preprocessing [
8
], feature
extraction [
9
] and classification methods [
10
] typically require substantial expertise and
a priori knowledge about the BCI paradigm. Moreover, due to manual processing, some
Brain Sci. 2023, 13, 268. https://doi.org/10.3390/brainsci13020268 https://www.mdpi.com/journal/brainsci
