Citation: Arı, E.; Taçgın, E. Input
Shape Effect on Classification
Performance of Raw EEG Motor
Imagery Signals with Convolutional
Neural Networks for Use in
Brain—Computer Interfaces. Brain
Sci. 2023, 13, 240. https://doi.org/
10.3390/brainsci13020240
Academic Editors: Enrico Vezzetti,
Andrea Luigi Guerra, Gabriele
Baronio, Domenico Speranza
and Luca Ulrich
Received: 1 January 2023
Revised: 26 January 2023
Accepted: 28 January 2023
Published: 31 January 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
Input Shape Effect on Classification Performance of Raw EEG
Motor Imagery Signals with Convolutional Neural Networks
for Use in Brain—Computer Interfaces
Emre Arı
1,2,
* and Ertu˘grul Taçgın
3
1
Department of Mechanical Engineering, Faculty of Engineering, Marmara University, Istanbul 34840, Turkey
2
Department of Mechanical Engineering, Faculty of Engineering, Dicle University, Diyarbakır 21280, Turkey
3
Department of Mechanical Engineering, Faculty of Engineering, Do˘gu¸s University, Istanbul 34775, Turkey
* Correspondence: emreari@marun.edu.tr
Abstract:
EEG signals are interpreted, analyzed and classified by many researchers for use in
brain–computer interfaces. Although there are many different EEG signal acquisition methods,
one of the most interesting is motor imagery signals. Many different signal processing methods,
machine learning and deep learning models have been developed for the classification of motor im-
agery signals. Among these, Convolutional Neural Network models generally achieve better results
than other models. Because the size and shape of the data is important for training Convolutional
Neural Network models and discovering the right relationships, researchers have designed and
experimented with many different input shape structures. However, no study has been found in the
literature evaluating the effect of different input shapes on model performance and accuracy. In this
study, the effects of different input shapes on model performance and accuracy in the classification of
EEG motor imagery signals were investigated, which had not been specifically studied before. In
addition, signal preprocessing methods, which take a long time before classification, were not used;
rather, two CNN models were developed for training and classification using raw data. Two different
datasets, BCI Competition IV 2A and 2B, were used in classification processes. For different input
shapes, 53.03–89.29% classification accuracy and 2–23 s epoch time were obtained for 2A dataset,
64.84–84.94% classification accuracy and 4–10 s epoch time were obtained for 2B dataset. This study
showed that the input shape has a significant effect on the classification performance, and when the
correct input shape is selected and the correct CNN architecture is developed, feature extraction and
classification can be done well by the CNN architecture without any signal preprocessing.
Keywords:
brain–computer interface (BCI); deep learning; EEG motor imagery; classification; input
shape; raw data
1. Introduction
Brain–computer interfaces enable the use of an external device by collecting and
processing brain signals in various online or offline methods. To collect brain signals,
electrocorticograms (ECoG), magnetoencephalography (MEG), electroencephalography
(EEG), positron emission topography (PET), local field potentials and action potentials,
functional magnetic resonance imaging (fMRI), and near-infrared spectral imaging (NIRS)
methods are used. The distinctive features of these collected signals are obtained by
different signal processing techniques. These distinguishing features are classified and
integrated into a control system or used as a control signal by connecting to a device
provided to fulfill the desired purpose.
Brain–computer interfaces are intensively researched in the fields of disease detection,
entertainment, education, marketing, games, medical devices and equipment, robotics
and physiotherapy, and real-time applications have tried to be developed [
1
–
13
]. One
of the most preferred signal types for brain–computer interfaces is EEG signals. It is
Brain Sci. 2023, 13, 240. https://doi.org/10.3390/brainsci13020240 https://www.mdpi.com/journal/brainsci
