Loading…
A Deep Transfer Convolutional Neural Network Framework for EEG Signal Classification
Nowadays, motor imagery (MI) electroencephalogram (EEG) signal classification has become a hotspot in the research field of brain computer interface (BCI). More recently, deep learning has emerged as a promising technique to automatically extract features of raw MI EEG signals and then classify them...
Saved in:
Published in: | IEEE access 2019, Vol.7, p.112767-112776 |
---|---|
Main Authors: | , , , , , , , , |
Format: | Article |
Language: | English |
Subjects: | |
Citations: | Items that this one cites Items that cite this one |
Online Access: | Get full text |
Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
cited_by | cdi_FETCH-LOGICAL-c524t-9dc6ef5ffa48364fa468b93723581193f2903c6b9cfd2cf461c756b2c6f0dafd3 |
---|---|
cites | cdi_FETCH-LOGICAL-c524t-9dc6ef5ffa48364fa468b93723581193f2903c6b9cfd2cf461c756b2c6f0dafd3 |
container_end_page | 112776 |
container_issue | |
container_start_page | 112767 |
container_title | IEEE access |
container_volume | 7 |
creator | Xu, Gaowei Shen, Xiaoang Chen, Sirui Zong, Yongshuo Zhang, Canyang Yue, Hongyang Liu, Min Chen, Fei Che, Wenliang |
description | Nowadays, motor imagery (MI) electroencephalogram (EEG) signal classification has become a hotspot in the research field of brain computer interface (BCI). More recently, deep learning has emerged as a promising technique to automatically extract features of raw MI EEG signals and then classify them. However, deep learning-based methods still face two challenging problems in practical MI EEG signal classification applications: (1) Generally, training a deep learning model successfully needs a large amount of labeled data. However, most of the EEG signal data is unlabeled and it is quite difficult or even impossible for human experts to label all the signal samples manually. (2) It is extremely time-consuming and computationally expensive to train a deep learning model from scratch. To cope with these two challenges, a deep transfer convolutional neural network (CNN) framework based on VGG-16 is proposed for EEG signal classification. The proposed framework consists of a VGG-16 CNN model pre-trained on the ImageNet and a target CNN model which shares the same structure with VGG-16 except for the softmax output layer. The parameters of the pre-trained VGG-16 CNN model are directly transferred to the target CNN model used for MI EEG signal classification. Then, front-layers parameters in the target model are frozen, while later-layers parameters are fine-tuned by the target MI dataset. The target dataset is composed of time-frequency spectrum images of EEG signals. The performance of the proposed framework is verified on the public benchmark dataset 2b from the BCI competition IV. The experimental results show that the proposed framework improves the accuracy and efficiency performance of EEG signal classification compared with traditional methods, including support vector machine (SVM), artificial neural network (ANN), and standard CNN. |
doi_str_mv | 10.1109/ACCESS.2019.2930958 |
format | article |
fullrecord | <record><control><sourceid>proquest_ieee_</sourceid><recordid>TN_cdi_ieee_primary_8772136</recordid><sourceformat>XML</sourceformat><sourcesystem>PC</sourcesystem><ieee_id>8772136</ieee_id><doaj_id>oai_doaj_org_article_3c85f2d607bd4e1ba67e004cd51b63a3</doaj_id><sourcerecordid>2455608952</sourcerecordid><originalsourceid>FETCH-LOGICAL-c524t-9dc6ef5ffa48364fa468b93723581193f2903c6b9cfd2cf461c756b2c6f0dafd3</originalsourceid><addsrcrecordid>eNpNUctOwzAQtBBIVKVf0Eskzil-xE58rEJaKlVwaDlbjh-VS1oXOwXx9yQNqtjLrEYzs1oNAFMEZwhB_jQvy2qzmWGI-AxzAjktbsAII8ZTQgm7_bffg0mMe9hN0VE0H4HtPHk25pRsgzxGa0JS-uOXb86t80fZJK_mHC7QfvvwkSyCPJjLZn1IqmqZbNyu15WNjNFZp2RvfAB3VjbRTP5wDN4X1bZ8Sddvy1U5X6eK4qxNuVbMWGqtzArCsg5YUXOSY0ILhDixmEOiWM2V1VjZjCGVU1ZjxSzU0moyBqshV3u5F6fgDjL8CC-duBA-7IQMrVONEUQV1GLNYF7rzKBastxAmClNUc2IJF3W45B1Cv7zbGIr9v4cut-iwBmlDBac4k5FBpUKPsZg7PUqgqJvQwxtiL4N8ddG55oOLmeMuTqKPMeIMPILnwGFgw</addsrcrecordid><sourcetype>Open Website</sourcetype><iscdi>true</iscdi><recordtype>article</recordtype><pqid>2455608952</pqid></control><display><type>article</type><title>A Deep Transfer Convolutional Neural Network Framework for EEG Signal Classification</title><source>IEEE Xplore Open Access Journals</source><creator>Xu, Gaowei ; Shen, Xiaoang ; Chen, Sirui ; Zong, Yongshuo ; Zhang, Canyang ; Yue, Hongyang ; Liu, Min ; Chen, Fei ; Che, Wenliang</creator><creatorcontrib>Xu, Gaowei ; Shen, Xiaoang ; Chen, Sirui ; Zong, Yongshuo ; Zhang, Canyang ; Yue, Hongyang ; Liu, Min ; Chen, Fei ; Che, Wenliang</creatorcontrib><description>Nowadays, motor imagery (MI) electroencephalogram (EEG) signal classification has become a hotspot in the research field of brain computer interface (BCI). More recently, deep learning has emerged as a promising technique to automatically extract features of raw MI EEG signals and then classify them. However, deep learning-based methods still face two challenging problems in practical MI EEG signal classification applications: (1) Generally, training a deep learning model successfully needs a large amount of labeled data. However, most of the EEG signal data is unlabeled and it is quite difficult or even impossible for human experts to label all the signal samples manually. (2) It is extremely time-consuming and computationally expensive to train a deep learning model from scratch. To cope with these two challenges, a deep transfer convolutional neural network (CNN) framework based on VGG-16 is proposed for EEG signal classification. The proposed framework consists of a VGG-16 CNN model pre-trained on the ImageNet and a target CNN model which shares the same structure with VGG-16 except for the softmax output layer. The parameters of the pre-trained VGG-16 CNN model are directly transferred to the target CNN model used for MI EEG signal classification. Then, front-layers parameters in the target model are frozen, while later-layers parameters are fine-tuned by the target MI dataset. The target dataset is composed of time-frequency spectrum images of EEG signals. The performance of the proposed framework is verified on the public benchmark dataset 2b from the BCI competition IV. The experimental results show that the proposed framework improves the accuracy and efficiency performance of EEG signal classification compared with traditional methods, including support vector machine (SVM), artificial neural network (ANN), and standard CNN.</description><identifier>ISSN: 2169-3536</identifier><identifier>EISSN: 2169-3536</identifier><identifier>DOI: 10.1109/ACCESS.2019.2930958</identifier><identifier>CODEN: IAECCG</identifier><language>eng</language><publisher>Piscataway: IEEE</publisher><subject>Artificial neural networks ; Brain modeling ; Classification ; Computational modeling ; Datasets ; Deep learning ; electroencephalogram (EEG) ; Electroencephalography ; Feature extraction ; Frequency spectrum ; Human-computer interface ; Image classification ; Learning theory ; Machine learning ; Mathematical models ; Motor imagery (MI) ; Neural networks ; Parameters ; short time Fourier transform (STFT) ; Signal classification ; Support vector machines ; Target recognition ; Task analysis ; Time-frequency analysis ; transfer learning ; VGG-16</subject><ispartof>IEEE access, 2019, Vol.7, p.112767-112776</ispartof><rights>Copyright The Institute of Electrical and Electronics Engineers, Inc. (IEEE) 2019</rights><lds50>peer_reviewed</lds50><oa>free_for_read</oa><woscitedreferencessubscribed>false</woscitedreferencessubscribed><citedby>FETCH-LOGICAL-c524t-9dc6ef5ffa48364fa468b93723581193f2903c6b9cfd2cf461c756b2c6f0dafd3</citedby><cites>FETCH-LOGICAL-c524t-9dc6ef5ffa48364fa468b93723581193f2903c6b9cfd2cf461c756b2c6f0dafd3</cites><orcidid>0000-0002-8902-5460 ; 0000-0003-3752-7749</orcidid></display><links><openurl>$$Topenurl_article</openurl><openurlfulltext>$$Topenurlfull_article</openurlfulltext><thumbnail>$$Tsyndetics_thumb_exl</thumbnail><linktohtml>$$Uhttps://ieeexplore.ieee.org/document/8772136$$EHTML$$P50$$Gieee$$Hfree_for_read</linktohtml><link.rule.ids>314,780,784,4024,27633,27923,27924,27925,54933</link.rule.ids></links><search><creatorcontrib>Xu, Gaowei</creatorcontrib><creatorcontrib>Shen, Xiaoang</creatorcontrib><creatorcontrib>Chen, Sirui</creatorcontrib><creatorcontrib>Zong, Yongshuo</creatorcontrib><creatorcontrib>Zhang, Canyang</creatorcontrib><creatorcontrib>Yue, Hongyang</creatorcontrib><creatorcontrib>Liu, Min</creatorcontrib><creatorcontrib>Chen, Fei</creatorcontrib><creatorcontrib>Che, Wenliang</creatorcontrib><title>A Deep Transfer Convolutional Neural Network Framework for EEG Signal Classification</title><title>IEEE access</title><addtitle>Access</addtitle><description>Nowadays, motor imagery (MI) electroencephalogram (EEG) signal classification has become a hotspot in the research field of brain computer interface (BCI). More recently, deep learning has emerged as a promising technique to automatically extract features of raw MI EEG signals and then classify them. However, deep learning-based methods still face two challenging problems in practical MI EEG signal classification applications: (1) Generally, training a deep learning model successfully needs a large amount of labeled data. However, most of the EEG signal data is unlabeled and it is quite difficult or even impossible for human experts to label all the signal samples manually. (2) It is extremely time-consuming and computationally expensive to train a deep learning model from scratch. To cope with these two challenges, a deep transfer convolutional neural network (CNN) framework based on VGG-16 is proposed for EEG signal classification. The proposed framework consists of a VGG-16 CNN model pre-trained on the ImageNet and a target CNN model which shares the same structure with VGG-16 except for the softmax output layer. The parameters of the pre-trained VGG-16 CNN model are directly transferred to the target CNN model used for MI EEG signal classification. Then, front-layers parameters in the target model are frozen, while later-layers parameters are fine-tuned by the target MI dataset. The target dataset is composed of time-frequency spectrum images of EEG signals. The performance of the proposed framework is verified on the public benchmark dataset 2b from the BCI competition IV. The experimental results show that the proposed framework improves the accuracy and efficiency performance of EEG signal classification compared with traditional methods, including support vector machine (SVM), artificial neural network (ANN), and standard CNN.</description><subject>Artificial neural networks</subject><subject>Brain modeling</subject><subject>Classification</subject><subject>Computational modeling</subject><subject>Datasets</subject><subject>Deep learning</subject><subject>electroencephalogram (EEG)</subject><subject>Electroencephalography</subject><subject>Feature extraction</subject><subject>Frequency spectrum</subject><subject>Human-computer interface</subject><subject>Image classification</subject><subject>Learning theory</subject><subject>Machine learning</subject><subject>Mathematical models</subject><subject>Motor imagery (MI)</subject><subject>Neural networks</subject><subject>Parameters</subject><subject>short time Fourier transform (STFT)</subject><subject>Signal classification</subject><subject>Support vector machines</subject><subject>Target recognition</subject><subject>Task analysis</subject><subject>Time-frequency analysis</subject><subject>transfer learning</subject><subject>VGG-16</subject><issn>2169-3536</issn><issn>2169-3536</issn><fulltext>true</fulltext><rsrctype>article</rsrctype><creationdate>2019</creationdate><recordtype>article</recordtype><sourceid>ESBDL</sourceid><sourceid>DOA</sourceid><recordid>eNpNUctOwzAQtBBIVKVf0Eskzil-xE58rEJaKlVwaDlbjh-VS1oXOwXx9yQNqtjLrEYzs1oNAFMEZwhB_jQvy2qzmWGI-AxzAjktbsAII8ZTQgm7_bffg0mMe9hN0VE0H4HtPHk25pRsgzxGa0JS-uOXb86t80fZJK_mHC7QfvvwkSyCPJjLZn1IqmqZbNyu15WNjNFZp2RvfAB3VjbRTP5wDN4X1bZ8Sddvy1U5X6eK4qxNuVbMWGqtzArCsg5YUXOSY0ILhDixmEOiWM2V1VjZjCGVU1ZjxSzU0moyBqshV3u5F6fgDjL8CC-duBA-7IQMrVONEUQV1GLNYF7rzKBastxAmClNUc2IJF3W45B1Cv7zbGIr9v4cut-iwBmlDBac4k5FBpUKPsZg7PUqgqJvQwxtiL4N8ddG55oOLmeMuTqKPMeIMPILnwGFgw</recordid><startdate>2019</startdate><enddate>2019</enddate><creator>Xu, Gaowei</creator><creator>Shen, Xiaoang</creator><creator>Chen, Sirui</creator><creator>Zong, Yongshuo</creator><creator>Zhang, Canyang</creator><creator>Yue, Hongyang</creator><creator>Liu, Min</creator><creator>Chen, Fei</creator><creator>Che, Wenliang</creator><general>IEEE</general><general>The Institute of Electrical and Electronics Engineers, Inc. (IEEE)</general><scope>97E</scope><scope>ESBDL</scope><scope>RIA</scope><scope>RIE</scope><scope>AAYXX</scope><scope>CITATION</scope><scope>7SC</scope><scope>7SP</scope><scope>7SR</scope><scope>8BQ</scope><scope>8FD</scope><scope>JG9</scope><scope>JQ2</scope><scope>L7M</scope><scope>L~C</scope><scope>L~D</scope><scope>DOA</scope><orcidid>https://orcid.org/0000-0002-8902-5460</orcidid><orcidid>https://orcid.org/0000-0003-3752-7749</orcidid></search><sort><creationdate>2019</creationdate><title>A Deep Transfer Convolutional Neural Network Framework for EEG Signal Classification</title><author>Xu, Gaowei ; Shen, Xiaoang ; Chen, Sirui ; Zong, Yongshuo ; Zhang, Canyang ; Yue, Hongyang ; Liu, Min ; Chen, Fei ; Che, Wenliang</author></sort><facets><frbrtype>5</frbrtype><frbrgroupid>cdi_FETCH-LOGICAL-c524t-9dc6ef5ffa48364fa468b93723581193f2903c6b9cfd2cf461c756b2c6f0dafd3</frbrgroupid><rsrctype>articles</rsrctype><prefilter>articles</prefilter><language>eng</language><creationdate>2019</creationdate><topic>Artificial neural networks</topic><topic>Brain modeling</topic><topic>Classification</topic><topic>Computational modeling</topic><topic>Datasets</topic><topic>Deep learning</topic><topic>electroencephalogram (EEG)</topic><topic>Electroencephalography</topic><topic>Feature extraction</topic><topic>Frequency spectrum</topic><topic>Human-computer interface</topic><topic>Image classification</topic><topic>Learning theory</topic><topic>Machine learning</topic><topic>Mathematical models</topic><topic>Motor imagery (MI)</topic><topic>Neural networks</topic><topic>Parameters</topic><topic>short time Fourier transform (STFT)</topic><topic>Signal classification</topic><topic>Support vector machines</topic><topic>Target recognition</topic><topic>Task analysis</topic><topic>Time-frequency analysis</topic><topic>transfer learning</topic><topic>VGG-16</topic><toplevel>peer_reviewed</toplevel><toplevel>online_resources</toplevel><creatorcontrib>Xu, Gaowei</creatorcontrib><creatorcontrib>Shen, Xiaoang</creatorcontrib><creatorcontrib>Chen, Sirui</creatorcontrib><creatorcontrib>Zong, Yongshuo</creatorcontrib><creatorcontrib>Zhang, Canyang</creatorcontrib><creatorcontrib>Yue, Hongyang</creatorcontrib><creatorcontrib>Liu, Min</creatorcontrib><creatorcontrib>Chen, Fei</creatorcontrib><creatorcontrib>Che, Wenliang</creatorcontrib><collection>IEEE All-Society Periodicals Package (ASPP) 2005-present</collection><collection>IEEE Xplore Open Access Journals</collection><collection>IEEE All-Society Periodicals Package (ASPP) 1998–Present</collection><collection>IEEE Electronic Library Online</collection><collection>CrossRef</collection><collection>Computer and Information Systems Abstracts</collection><collection>Electronics & Communications Abstracts</collection><collection>Engineered Materials Abstracts</collection><collection>METADEX</collection><collection>Technology Research Database</collection><collection>Materials Research Database</collection><collection>ProQuest Computer Science Collection</collection><collection>Advanced Technologies Database with Aerospace</collection><collection>Computer and Information Systems Abstracts Academic</collection><collection>Computer and Information Systems Abstracts Professional</collection><collection>DOAJ Directory of Open Access Journals</collection><jtitle>IEEE access</jtitle></facets><delivery><delcategory>Remote Search Resource</delcategory><fulltext>fulltext</fulltext></delivery><addata><au>Xu, Gaowei</au><au>Shen, Xiaoang</au><au>Chen, Sirui</au><au>Zong, Yongshuo</au><au>Zhang, Canyang</au><au>Yue, Hongyang</au><au>Liu, Min</au><au>Chen, Fei</au><au>Che, Wenliang</au><format>journal</format><genre>article</genre><ristype>JOUR</ristype><atitle>A Deep Transfer Convolutional Neural Network Framework for EEG Signal Classification</atitle><jtitle>IEEE access</jtitle><stitle>Access</stitle><date>2019</date><risdate>2019</risdate><volume>7</volume><spage>112767</spage><epage>112776</epage><pages>112767-112776</pages><issn>2169-3536</issn><eissn>2169-3536</eissn><coden>IAECCG</coden><abstract>Nowadays, motor imagery (MI) electroencephalogram (EEG) signal classification has become a hotspot in the research field of brain computer interface (BCI). More recently, deep learning has emerged as a promising technique to automatically extract features of raw MI EEG signals and then classify them. However, deep learning-based methods still face two challenging problems in practical MI EEG signal classification applications: (1) Generally, training a deep learning model successfully needs a large amount of labeled data. However, most of the EEG signal data is unlabeled and it is quite difficult or even impossible for human experts to label all the signal samples manually. (2) It is extremely time-consuming and computationally expensive to train a deep learning model from scratch. To cope with these two challenges, a deep transfer convolutional neural network (CNN) framework based on VGG-16 is proposed for EEG signal classification. The proposed framework consists of a VGG-16 CNN model pre-trained on the ImageNet and a target CNN model which shares the same structure with VGG-16 except for the softmax output layer. The parameters of the pre-trained VGG-16 CNN model are directly transferred to the target CNN model used for MI EEG signal classification. Then, front-layers parameters in the target model are frozen, while later-layers parameters are fine-tuned by the target MI dataset. The target dataset is composed of time-frequency spectrum images of EEG signals. The performance of the proposed framework is verified on the public benchmark dataset 2b from the BCI competition IV. The experimental results show that the proposed framework improves the accuracy and efficiency performance of EEG signal classification compared with traditional methods, including support vector machine (SVM), artificial neural network (ANN), and standard CNN.</abstract><cop>Piscataway</cop><pub>IEEE</pub><doi>10.1109/ACCESS.2019.2930958</doi><tpages>10</tpages><orcidid>https://orcid.org/0000-0002-8902-5460</orcidid><orcidid>https://orcid.org/0000-0003-3752-7749</orcidid><oa>free_for_read</oa></addata></record> |
fulltext | fulltext |
identifier | ISSN: 2169-3536 |
ispartof | IEEE access, 2019, Vol.7, p.112767-112776 |
issn | 2169-3536 2169-3536 |
language | eng |
recordid | cdi_ieee_primary_8772136 |
source | IEEE Xplore Open Access Journals |
subjects | Artificial neural networks Brain modeling Classification Computational modeling Datasets Deep learning electroencephalogram (EEG) Electroencephalography Feature extraction Frequency spectrum Human-computer interface Image classification Learning theory Machine learning Mathematical models Motor imagery (MI) Neural networks Parameters short time Fourier transform (STFT) Signal classification Support vector machines Target recognition Task analysis Time-frequency analysis transfer learning VGG-16 |
title | A Deep Transfer Convolutional Neural Network Framework for EEG Signal Classification |
url | http://sfxeu10.hosted.exlibrisgroup.com/loughborough?ctx_ver=Z39.88-2004&ctx_enc=info:ofi/enc:UTF-8&ctx_tim=2024-12-25T12%3A35%3A39IST&url_ver=Z39.88-2004&url_ctx_fmt=infofi/fmt:kev:mtx:ctx&rfr_id=info:sid/primo.exlibrisgroup.com:primo3-Article-proquest_ieee_&rft_val_fmt=info:ofi/fmt:kev:mtx:journal&rft.genre=article&rft.atitle=A%20Deep%20Transfer%20Convolutional%20Neural%20Network%20Framework%20for%20EEG%20Signal%20Classification&rft.jtitle=IEEE%20access&rft.au=Xu,%20Gaowei&rft.date=2019&rft.volume=7&rft.spage=112767&rft.epage=112776&rft.pages=112767-112776&rft.issn=2169-3536&rft.eissn=2169-3536&rft.coden=IAECCG&rft_id=info:doi/10.1109/ACCESS.2019.2930958&rft_dat=%3Cproquest_ieee_%3E2455608952%3C/proquest_ieee_%3E%3Cgrp_id%3Ecdi_FETCH-LOGICAL-c524t-9dc6ef5ffa48364fa468b93723581193f2903c6b9cfd2cf461c756b2c6f0dafd3%3C/grp_id%3E%3Coa%3E%3C/oa%3E%3Curl%3E%3C/url%3E&rft_id=info:oai/&rft_pqid=2455608952&rft_id=info:pmid/&rft_ieee_id=8772136&rfr_iscdi=true |