A systematic comparison of deep learning methods for EEG time series analysis
Analyzing time series data like EEG or MEG is challenging due to noisy, high-dimensional, and patient-specific signals. Deep learning methods have been demonstrated to be superior in analyzing time series data compared to shallow learning methods which utilize handcrafted and often subjective featur...
Main Authors: | , , , |
---|---|
Format: | Article |
Language: | English |
Published: |
Frontiers Media S.A.
2023-02-01
|
Series: | Frontiers in Neuroinformatics |
Subjects: | |
Online Access: | https://www.frontiersin.org/articles/10.3389/fninf.2023.1067095/full |
_version_ | 1797897106248171520 |
---|---|
author | Dominik Walther Johannes Viehweg Jens Haueisen Patrick Mäder Patrick Mäder |
author_facet | Dominik Walther Johannes Viehweg Jens Haueisen Patrick Mäder Patrick Mäder |
author_sort | Dominik Walther |
collection | DOAJ |
description | Analyzing time series data like EEG or MEG is challenging due to noisy, high-dimensional, and patient-specific signals. Deep learning methods have been demonstrated to be superior in analyzing time series data compared to shallow learning methods which utilize handcrafted and often subjective features. Especially, recurrent deep neural networks (RNN) are considered suitable to analyze such continuous data. However, previous studies show that they are computationally expensive and difficult to train. In contrast, feed-forward networks (FFN) have previously mostly been considered in combination with hand-crafted and problem-specific feature extractions, such as short time Fourier and discrete wavelet transform. A sought-after are easily applicable methods that efficiently analyze raw data to remove the need for problem-specific adaptations. In this work, we systematically compare RNN and FFN topologies as well as advanced architectural concepts on multiple datasets with the same data preprocessing pipeline. We examine the behavior of those approaches to provide an update and guideline for researchers who deal with automated analysis of EEG time series data. To ensure that the results are meaningful, it is important to compare the presented approaches while keeping the same experimental setup, which to our knowledge was never done before. This paper is a first step toward a fairer comparison of different methodologies with EEG time series data. Our results indicate that a recurrent LSTM architecture with attention performs best on less complex tasks, while the temporal convolutional network (TCN) outperforms all the recurrent architectures on the most complex dataset yielding a 8.61% accuracy improvement. In general, we found the attention mechanism to substantially improve classification results of RNNs. Toward a light-weight and online learning-ready approach, we found extreme learning machines (ELM) to yield comparable results for the less complex tasks. |
first_indexed | 2024-04-10T07:53:34Z |
format | Article |
id | doaj.art-d89475568a6b4bff96af59fe641feb3f |
institution | Directory Open Access Journal |
issn | 1662-5196 |
language | English |
last_indexed | 2024-04-10T07:53:34Z |
publishDate | 2023-02-01 |
publisher | Frontiers Media S.A. |
record_format | Article |
series | Frontiers in Neuroinformatics |
spelling | doaj.art-d89475568a6b4bff96af59fe641feb3f2023-02-23T07:44:00ZengFrontiers Media S.A.Frontiers in Neuroinformatics1662-51962023-02-011710.3389/fninf.2023.10670951067095A systematic comparison of deep learning methods for EEG time series analysisDominik Walther0Johannes Viehweg1Jens Haueisen2Patrick Mäder3Patrick Mäder4Data-Intensive Systems and Visualization Group (dAI.SY), Technische Universität Ilmenau, Ilmenau, GermanyData-Intensive Systems and Visualization Group (dAI.SY), Technische Universität Ilmenau, Ilmenau, GermanyInstitute of Biomedical Engineering and Informatics, Technische Universität Ilmenau, Ilmenau, GermanyData-Intensive Systems and Visualization Group (dAI.SY), Technische Universität Ilmenau, Ilmenau, GermanyFaculty of Biological Sciences, Friedrich Schiller University, Jena, GermanyAnalyzing time series data like EEG or MEG is challenging due to noisy, high-dimensional, and patient-specific signals. Deep learning methods have been demonstrated to be superior in analyzing time series data compared to shallow learning methods which utilize handcrafted and often subjective features. Especially, recurrent deep neural networks (RNN) are considered suitable to analyze such continuous data. However, previous studies show that they are computationally expensive and difficult to train. In contrast, feed-forward networks (FFN) have previously mostly been considered in combination with hand-crafted and problem-specific feature extractions, such as short time Fourier and discrete wavelet transform. A sought-after are easily applicable methods that efficiently analyze raw data to remove the need for problem-specific adaptations. In this work, we systematically compare RNN and FFN topologies as well as advanced architectural concepts on multiple datasets with the same data preprocessing pipeline. We examine the behavior of those approaches to provide an update and guideline for researchers who deal with automated analysis of EEG time series data. To ensure that the results are meaningful, it is important to compare the presented approaches while keeping the same experimental setup, which to our knowledge was never done before. This paper is a first step toward a fairer comparison of different methodologies with EEG time series data. Our results indicate that a recurrent LSTM architecture with attention performs best on less complex tasks, while the temporal convolutional network (TCN) outperforms all the recurrent architectures on the most complex dataset yielding a 8.61% accuracy improvement. In general, we found the attention mechanism to substantially improve classification results of RNNs. Toward a light-weight and online learning-ready approach, we found extreme learning machines (ELM) to yield comparable results for the less complex tasks.https://www.frontiersin.org/articles/10.3389/fninf.2023.1067095/fullrecurrent neural networksfeed forward neural networkstime series analysisattentiontransformer networks |
spellingShingle | Dominik Walther Johannes Viehweg Jens Haueisen Patrick Mäder Patrick Mäder A systematic comparison of deep learning methods for EEG time series analysis Frontiers in Neuroinformatics recurrent neural networks feed forward neural networks time series analysis attention transformer networks |
title | A systematic comparison of deep learning methods for EEG time series analysis |
title_full | A systematic comparison of deep learning methods for EEG time series analysis |
title_fullStr | A systematic comparison of deep learning methods for EEG time series analysis |
title_full_unstemmed | A systematic comparison of deep learning methods for EEG time series analysis |
title_short | A systematic comparison of deep learning methods for EEG time series analysis |
title_sort | systematic comparison of deep learning methods for eeg time series analysis |
topic | recurrent neural networks feed forward neural networks time series analysis attention transformer networks |
url | https://www.frontiersin.org/articles/10.3389/fninf.2023.1067095/full |
work_keys_str_mv | AT dominikwalther asystematiccomparisonofdeeplearningmethodsforeegtimeseriesanalysis AT johannesviehweg asystematiccomparisonofdeeplearningmethodsforeegtimeseriesanalysis AT jenshaueisen asystematiccomparisonofdeeplearningmethodsforeegtimeseriesanalysis AT patrickmader asystematiccomparisonofdeeplearningmethodsforeegtimeseriesanalysis AT patrickmader asystematiccomparisonofdeeplearningmethodsforeegtimeseriesanalysis AT dominikwalther systematiccomparisonofdeeplearningmethodsforeegtimeseriesanalysis AT johannesviehweg systematiccomparisonofdeeplearningmethodsforeegtimeseriesanalysis AT jenshaueisen systematiccomparisonofdeeplearningmethodsforeegtimeseriesanalysis AT patrickmader systematiccomparisonofdeeplearningmethodsforeegtimeseriesanalysis AT patrickmader systematiccomparisonofdeeplearningmethodsforeegtimeseriesanalysis |