Please use this identifier to cite or link to this item: http://bura.brunel.ac.uk/handle/2438/30315
Title: Spatial Variation Generation Algorithm for Motor Imagery Data Augmentation: Increasing the Density of Sample Vicinity
Authors: Qin, C
Yang, R
Huang, M
Liu, W
Wang, Z
Keywords: brain–computer interfaces;data augmentation;deep learning;electroencephalogram
Issue Date: 12-Sep-2024
Publisher: Institute of Electrical and Electronics Engineers (IEEE)
Citation: Qin, C. et al. (2024) 'Spatial Variation Generation Algorithm for Motor Imagery Data Augmentation: Increasing the Density of Sample Vicinity', IEEE Transactions on Neural Systems and Rehabilitation Engineering, 31, pp. 3675 - 3686. doi: 10.1109/TNSRE.2023.3314679.
Abstract: The imbalanced development between deep learning-based model design and motor imagery (MI) data acquisition raises concerns about the potential overfitting issue—models can identify training data well but fail to generalize test data. In this study, a Spatial Variation Generation (SVG) algorithm for MI data augmentation is proposed to alleviate the overfitting issue. In essence, SVG generates MI data using variations of electrode placement and brain spatial pattern, ultimately elevating the density of the raw sample vicinity. The proposed SVG prevents models from memorizing the training data by replacing the raw samples with the proper vicinal distribution. Moreover, SVG generates a uniform distribution and stabilizes the training process of models. In comparison studies involving five deep learning-based models across eight datasets, the proposed SVG algorithm exhibited a notable improvement of 0.021 in the area under the receiver operating characteristic curve (AUC). The improvement achieved by SVG outperforms other data augmentation algorithms. Further results from the ablation study verify the effectiveness of each component of SVG. Finally, the studies in the control group with varying numbers of samples show that the SVG algorithm consistently improves the AUC, with improvements ranging from approximately 0.02 to 0.15.
URI: https://bura.brunel.ac.uk/handle/2438/30315
DOI: https://doi.org/10.1109/TNSRE.2023.3314679
ISSN: 1534-4320
Other Identifiers: ORCiD: Chengxuan Qin https://orcid.org/0009-0009-8463-3457
ORCiD: Rui Yang https://orcid.org/0000-0002-5634-5476
ORCiD: Mengjie Huang https://orcid.org/0000-0001-8163-8679
ORCiD: Weibo Liu https://orcid.org/0000-0002-8169-3261
ORCiD: Zidong Wang https://orcid.org/0000-0002-9576-7401
Appears in Collections:Dept of Computer Science Research Papers

Files in This Item:
File Description SizeFormat 
FullText.pdfCopyright © 2023 The Author(s). This work is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License. For more information, see https://creativecommons.org/licenses/by-nc-nd/4.0/10.04 MBAdobe PDFView/Open


This item is licensed under a Creative Commons License Creative Commons