CoChat

Data augmentation for motor imagery signal classification based on a hybrid neural network.

Open access
Zhang, Kai, Xu, Guanghua, Han, Zezhen, Ma, Kaiquan, Zheng, Xiaowei, Chen, Longting, Duan, Nan, Zhang, Sicong · 2020 · Sensors

The DOI matches its Crossref record; 74 of the 75 references it cites were found in a registry or an index.

Abstract

As an important paradigm of spontaneous brain-computer interfaces (BCIs), motor imagery (MI) has been widely used in the fields of neurological rehabilitation and robot control. Recently, researchers have proposed various methods for feature extraction and classification based on MI signals. The decoding model based on deep neural networks (DNNs) has attracted significant attention in the field of MI signal processing. Due to the strict requirements for subjects and experimental environments, it is difficult to collect large-scale and high-quality electroencephalogram (EEG) data. However, the performance of a deep learning model depends directly on the size of the datasets. Therefore, the decoding of MI-EEG signals based on a DNN has proven highly challenging in practice. Based on this, we investigated the performance of different data augmentation (DA) methods for the classification of MI data using a DNN. First, we transformed the time series signals into spectrogram images using a short-time Fourier transform (STFT). Then, we evaluated and compared the performance of different DA methods for this spectrogram data. Next, we developed a convolutional neural network (CNN) to classify the MI signals and compared the classification performance of after DA. The Fréchet inception distance (FID) was used to evaluate the quality of the generated data (GD) and the classification accuracy, and mean kappa values were used to explore the best CNN-DA method. In addition, analysis of variance (ANOVA) and paired t-tests were used to assess the significance of the results. The results showed that the deep convolutional generative adversarial network (DCGAN) provided better augmentation performance than traditional DA methods: geometric transformation (GT), autoencoder (AE), and variational autoencoder (VAE) (p < 0.01). Public datasets of the BCI competition IV (datasets 1 and 2b) were used to verify the classification performance. Improvements in the classification accuracies of 17% and 21% (p < 0.01) were observed after DA for the two datasets. In addition, the hybrid network CNN-DCGAN outperformed the other classification methods, with average kappa values of 0.564 and 0.677 for the two datasets.

Overview

The claims and statistics of this paper haven’t been checked yet.

Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.

A few free checks a day · papers already verified open instantly · results are public

Integrity signals

Registry

Crossref · 10.3390/s20164485

Cites

74 of 74 references verified0 citation errors · 0 not checkable

Statistics

Figures

Figure 1 Schematic diagram of the experiment and the timing during a session using the Brain–Computer Interface (BCI) Competition IV dataset 1.
Figure 1 Schematic diagram of the experiment and the timing during a session using the Brain–Computer Interface (BCI) Competition IV dataset 1.CC BY · PubMed Central
Figure 2 Schematic diagram of the experiment and the timing during a session using the BCI Competition IV dataset 2b.
Figure 2 Schematic diagram of the experiment and the timing during a session using the BCI Competition IV dataset 2b.CC BY · PubMed Central
Figure 3 Input images with the electrodes after the short-time Fourier transform (STFT).
Figure 3 Input images with the electrodes after the short-time Fourier transform (STFT).CC BY · PubMed Central
Figure 4 Data augmentation using geometric transformations.
Figure 4 Data augmentation using geometric transformations.CC BY · PubMed Central
Figure 5 Data augmentation using noise addition.
Figure 5 Data augmentation using noise addition.CC BY · PubMed Central
Figure 6 Data augmentation using an autoencoder.
Figure 6 Data augmentation using an autoencoder.CC BY · PubMed Central
Figure 7 Data augmentation using a variational autoencoder.
Figure 7 Data augmentation using a variational autoencoder.CC BY · PubMed Central
Figure 8 The structure of the generator.
Figure 8 The structure of the generator.CC BY · PubMed Central
Figure 9 The structure of the discriminator.
Figure 9 The structure of the discriminator.CC BY · PubMed Central
Figure 10 The structure of the deep convolutional generative adversarial network (DCGAN) model.
Figure 10 The structure of the deep convolutional generative adversarial network (DCGAN) model.CC BY · PubMed Central
Figure 11 The structure of the convolutional neural network (CNN) model.
Figure 11 The structure of the convolutional neural network (CNN) model.CC BY · PubMed Central
Figure 12 The ANOVA stats of the FID on data set 1 (a) and data set 2b (b).
Figure 12 The ANOVA stats of the FID on data set 1 (a) and data set 2b (b).CC BY · PubMed Central
Figure 13 The post hoc data after data augmentation using different methods.
Figure 13 The post hoc data after data augmentation using different methods.CC BY · PubMed Central
Figure 14 The ANOVA statistics of the classification accuracies of the methods for datasets 1 (a) and 2b (b).
Figure 14 The ANOVA statistics of the classification accuracies of the methods for datasets 1 (a) and 2b (b).CC BY · PubMed Central
Figure 15 The hybrid network for MI classification.
Figure 15 The hybrid network for MI classification.CC BY · PubMed Central

Verification

Claims the paper makes

The claims and statistics of this paper haven’t been checked yet.

Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.

A few free checks a day · papers already verified open instantly · results are public

Claims other papers make about this one

Each is checked against this paper’s own text: does it say what the citing paper says it does?

Self-check

Percentages, sums and reported differences re-computed from the paper’s own numbers.

The self-check needs the paper’s full text.

Citations

Cites

Show all 74 references

Research works printed but not linked: 3

  1. Logan E. Brandon T. Dimitris T. Ludwig S. Aleksander M. A rotation and a translation sufce: Fooling CNNs with simple transformations arXiv 2018 1712.02779
  2. Kingma D.P. Welling M. Auto-encoding variational bayes arXiv 2013 1312.6114
  3. Goodfellow I. Pouget-Abadie J. Mirza M. Xu B. Warde-Farely D. Ozair S. Courville A. Bengio Y. Generative Adversarial Nets Proceedings of the International Conference on Neural Information Processing Systems Montreal, QC, Canada 8–12 December 2014 2672 2680

Cited by

Metadata

Authors
Zhang, K., Xu, G., Han, Z., Ma, K., Zheng, X., Chen, L., Duan, N., & Zhang, S.
Affiliations
Xi'an Jiaotong University (CN)
Corresponding author
Xu, Guanghua
Venue
Sensors
Publisher
MDPI AG
Published
2020-08-11
Volume
20
Issue
16
Pages
4485
ISSN
1424-8220 (ISSN-L)
OpenAlex
W3048643622
Cited by
130
References
75
Citations by year
2026: 15 · 2025: 20 · 2024: 25 · 2023: 28 · 2022: 20 · 2021: 15 · 2020: 7
FWCI
9.40
Type
journal article
Language
en
Open access
Locations
  • Sensors (publishedVersion, cc-by)
  • PubMed Central (publishedVersion, cc-by)
  • PubMed (publishedVersion)
  • DOAJ (DOAJ: Directory of Open Access Journals) (submittedVersion, cc-by-sa)
Funders
  • National Natural Science Foundation of China (51775415)
  • National Key Research &amp; Development Plan of China (2017YFC1308500)
  • Key Research &amp; Development Plan of Shaanxi Province (2018ZDCXL-GY-06-01)
Topics
  • Life Sciences › Neuroscience › Cognitive Neuroscience › EEG and Brain-Computer Interfaces
  • Physical Sciences › Computer Science › Artificial Intelligence › Emotion and Mood Recognition
  • Physical Sciences › Engineering › Biomedical Engineering › Muscle activation and electromyography studies
Keywords
motor imagery, EEG classification, data augmentation, DCGAN, spectrogram images, short-time Fourier transform, convolutional neural networks, brain-computer interface, BCI Competition IV, hybrid neural network
MeSH terms
Algorithms, Electroencephalography, Humans, Imagination, Neural Networks, Computer, Brain-Computer Interfaces

About these checks

Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.

Statistics and arithmetic are recomputed, not judged.

Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.

What counts as a claim ›How we measure confidence ›How we check this ›

Are you an author, or do you see a problem with this page? Request a review

Data sources

Bibliographic data: Semantic Scholar, OpenAlex, Crossref.

Contains information from the Semantic Scholar Academic Graph Datasets, made available under the ODC Attribution License (ODC-BY 1.0).

Semantic Scholar’s data is described in: Kinney, R., et al. (2023). The Semantic Scholar Open Data Platform. arXiv:2301.10140.