Learning attentional temporal cues of brainwaves with spatial embedding for motion intent detection

Dalin Zhang; Kaixuan Chen; Debao Jian; Lina Yao; Sen Wang; Po Li

doi:10.1109/ICDM.2019.00189

Learning attentional temporal cues of brainwaves with spatial embedding for motion intent detection

Dalin Zhang, Kaixuan Chen, Debao Jian, Lina Yao, Sen Wang, Po Li

Research output: Contribution to book/anthology/report/conference proceeding › Article in proceeding › Research › peer-review

9 Citations (Scopus)

Abstract

As brain dynamics fluctuate considerably across different subjects, it is challenging to design effective handcrafted features based on prior knowledge. Regarding this gap, this paper proposes a Graph-based Convolutional Recurrent Attention Model (G-CRAM) to explore EEG features across different subjects for movement intention recognition. A graph structure is first developed to embed the positioning information of EEG nodes, and then a convolutional recurrent attention model learns EEG features from both spatial and temporal dimensions and adaptively emphasizes on the most distinguishable temporal periods. The proposed approach is validated on two public movement intention EEG datasets. The results show that the GCRAM achieves superior performance to state-of-the-art methods regarding recognition accuracy and ROC-AUC. Furthermore, model interpreting studies reveal the learning process of different neural network components and demonstrate that the proposed model can extract detailed features efficiently.

Original language	English
Title of host publication	Proceedings - 19th IEEE International Conference on Data Mining, ICDM 2019
Editors	Jianyong Wang, Kyuseok Shim, Xindong Wu
Number of pages	6
Publisher	IEEE
Publication date	Nov 2019
Pages	1450-1455
Article number	8970671
ISBN (Electronic)	9781728146034
DOIs	https://doi.org/10.1109/ICDM.2019.00189
Publication status	Published - Nov 2019
Externally published	Yes
Event	19th IEEE International Conference on Data Mining, ICDM 2019 - Beijing, China Duration: 8 Nov 2019 → 11 Nov 2019

Conference

Conference	19th IEEE International Conference on Data Mining, ICDM 2019
Country/Territory	China
City	Beijing
Period	08/11/2019 → 11/11/2019
Sponsor	Baidu, et al., IEEE Computer Society, LinkedIn, MiningLamp Technology., US National Science Foundation (NSF)

Series	Proceedings - IEEE International Conference on Data Mining, ICDM
Volume	2019-November
ISSN	1550-4786

Bibliographical note

Publisher Copyright:
© 2019 IEEE.

Keywords

Deep learning
EEG
Graph representation
Movement intention
Subject-independent

Access to Document

10.1109/ICDM.2019.00189

AUB Link

Search for the material in Aalborg University Library's search engine

Cite this

Zhang, Dalin ; Chen, Kaixuan ; Jian, Debao et al. / Learning attentional temporal cues of brainwaves with spatial embedding for motion intent detection. Proceedings - 19th IEEE International Conference on Data Mining, ICDM 2019. editor / Jianyong Wang ; Kyuseok Shim ; Xindong Wu. IEEE, 2019. pp. 1450-1455 (Proceedings - IEEE International Conference on Data Mining, ICDM, Vol. 2019-November).

@inproceedings{0bc1b37b9c2642208de52c24d48fda41,

title = "Learning attentional temporal cues of brainwaves with spatial embedding for motion intent detection",

abstract = "As brain dynamics fluctuate considerably across different subjects, it is challenging to design effective handcrafted features based on prior knowledge. Regarding this gap, this paper proposes a Graph-based Convolutional Recurrent Attention Model (G-CRAM) to explore EEG features across different subjects for movement intention recognition. A graph structure is first developed to embed the positioning information of EEG nodes, and then a convolutional recurrent attention model learns EEG features from both spatial and temporal dimensions and adaptively emphasizes on the most distinguishable temporal periods. The proposed approach is validated on two public movement intention EEG datasets. The results show that the GCRAM achieves superior performance to state-of-the-art methods regarding recognition accuracy and ROC-AUC. Furthermore, model interpreting studies reveal the learning process of different neural network components and demonstrate that the proposed model can extract detailed features efficiently.",

keywords = "Deep learning, EEG, Graph representation, Movement intention, Subject-independent",

author = "Dalin Zhang and Kaixuan Chen and Debao Jian and Lina Yao and Sen Wang and Po Li",

note = "Publisher Copyright: {\textcopyright} 2019 IEEE.; 19th IEEE International Conference on Data Mining, ICDM 2019 ; Conference date: 08-11-2019 Through 11-11-2019",

year = "2019",

month = nov,

doi = "10.1109/ICDM.2019.00189",

language = "English",

series = "Proceedings - IEEE International Conference on Data Mining, ICDM",

publisher = "IEEE",

pages = "1450--1455",

editor = "Jianyong Wang and Kyuseok Shim and Xindong Wu",

booktitle = "Proceedings - 19th IEEE International Conference on Data Mining, ICDM 2019",

address = "United States",

}

Zhang, D , Chen, K, Jian, D, Yao, L, Wang, S & Li, P 2019, Learning attentional temporal cues of brainwaves with spatial embedding for motion intent detection. in J Wang, K Shim & X Wu (eds), Proceedings - 19th IEEE International Conference on Data Mining, ICDM 2019., 8970671, IEEE, Proceedings - IEEE International Conference on Data Mining, ICDM, vol. 2019-November, pp. 1450-1455, 19th IEEE International Conference on Data Mining, ICDM 2019, Beijing, China, 08/11/2019. https://doi.org/10.1109/ICDM.2019.00189

Learning attentional temporal cues of brainwaves with spatial embedding for motion intent detection. / Zhang, Dalin ; Chen, Kaixuan; Jian, Debao et al.
Proceedings - 19th IEEE International Conference on Data Mining, ICDM 2019. ed. / Jianyong Wang; Kyuseok Shim; Xindong Wu. IEEE, 2019. p. 1450-1455 8970671 (Proceedings - IEEE International Conference on Data Mining, ICDM, Vol. 2019-November).

Research output: Contribution to book/anthology/report/conference proceeding › Article in proceeding › Research › peer-review

TY - GEN

T1 - Learning attentional temporal cues of brainwaves with spatial embedding for motion intent detection

AU - Zhang, Dalin

AU - Chen, Kaixuan

AU - Jian, Debao

AU - Yao, Lina

AU - Wang, Sen

AU - Li, Po

PY - 2019/11

Y1 - 2019/11

N2 - As brain dynamics fluctuate considerably across different subjects, it is challenging to design effective handcrafted features based on prior knowledge. Regarding this gap, this paper proposes a Graph-based Convolutional Recurrent Attention Model (G-CRAM) to explore EEG features across different subjects for movement intention recognition. A graph structure is first developed to embed the positioning information of EEG nodes, and then a convolutional recurrent attention model learns EEG features from both spatial and temporal dimensions and adaptively emphasizes on the most distinguishable temporal periods. The proposed approach is validated on two public movement intention EEG datasets. The results show that the GCRAM achieves superior performance to state-of-the-art methods regarding recognition accuracy and ROC-AUC. Furthermore, model interpreting studies reveal the learning process of different neural network components and demonstrate that the proposed model can extract detailed features efficiently.

AB - As brain dynamics fluctuate considerably across different subjects, it is challenging to design effective handcrafted features based on prior knowledge. Regarding this gap, this paper proposes a Graph-based Convolutional Recurrent Attention Model (G-CRAM) to explore EEG features across different subjects for movement intention recognition. A graph structure is first developed to embed the positioning information of EEG nodes, and then a convolutional recurrent attention model learns EEG features from both spatial and temporal dimensions and adaptively emphasizes on the most distinguishable temporal periods. The proposed approach is validated on two public movement intention EEG datasets. The results show that the GCRAM achieves superior performance to state-of-the-art methods regarding recognition accuracy and ROC-AUC. Furthermore, model interpreting studies reveal the learning process of different neural network components and demonstrate that the proposed model can extract detailed features efficiently.

KW - Deep learning

KW - EEG

KW - Graph representation

KW - Movement intention

KW - Subject-independent

UR - http://www.scopus.com/inward/record.url?scp=85078874843&partnerID=8YFLogxK

U2 - 10.1109/ICDM.2019.00189

DO - 10.1109/ICDM.2019.00189

M3 - Article in proceeding

AN - SCOPUS:85078874843

T3 - Proceedings - IEEE International Conference on Data Mining, ICDM

SP - 1450

EP - 1455

BT - Proceedings - 19th IEEE International Conference on Data Mining, ICDM 2019

A2 - Wang, Jianyong

A2 - Shim, Kyuseok

A2 - Wu, Xindong

PB - IEEE

T2 - 19th IEEE International Conference on Data Mining, ICDM 2019

Y2 - 8 November 2019 through 11 November 2019

ER -

Zhang D , Chen K, Jian D, Yao L, Wang S, Li P. Learning attentional temporal cues of brainwaves with spatial embedding for motion intent detection. In Wang J, Shim K, Wu X, editors, Proceedings - 19th IEEE International Conference on Data Mining, ICDM 2019. IEEE. 2019. p. 1450-1455. 8970671. (Proceedings - IEEE International Conference on Data Mining, ICDM, Vol. 2019-November). doi: 10.1109/ICDM.2019.00189

Learning attentional temporal cues of brainwaves with spatial embedding for motion intent detection

Abstract

Conference

Bibliographical note

Keywords

Access to Document

AUB Link

Other files and links

Fingerprint

Cite this