ConvNet-CA: A Lightweight Attention-Based CNN for Brain Disease Detection

Hengde Zhu; Jian Wang; Shui Hua Wang; Yu Dong Zhang; Juan M. Górriz

doi:10.1007/978-3-031-06242-1_1

ConvNet-CA: A Lightweight Attention-Based CNN for Brain Disease Detection

Hengde Zhu, Jian Wang, Shui Hua Wang, Yu Dong Zhang^*, Juan M. Górriz

^*Corresponding author for this work

Research output: Chapter in Book or Report/Conference proceeding › Conference Proceeding › peer-review

2 Citations (Scopus)

Abstract

Attention-based convolutional networks have attracted great interest in recent years and achieved great success in improving representation capability of networks. However, most attention mechanisms are complicated and implemented by introducing a large number of extra parameters. In this study, we proposed a lightweight attention-based convolutional network (ConvNet-CA) that has a low computation complexity yet a high performance for brain disease detection. ConvNet-CA weights the importance of different channels in features maps and pays more attention to important channels by introducing an efficient channel attention mechanism. We evaluated ConvNet-CA on a publicly accessible benchmark dataset: Whole Brain Atlas. The brain diseases involved in this study are stroke, neoplastic disease, degenerative disease, and infectious disease. The experimental results showed that ConvNet-CA achieved highly competitive performance over state-of-the-art methods on distinguishing different types of brain diseases, with an overall multi-class classification accuracy of 94.88 ± 3.64%.

Original language	English
Title of host publication	Artificial Intelligence in Neuroscience
Subtitle of host publication	Affective Analysis and Health Applications - 9th International Work-Conference on the Interplay Between Natural and Artificial Computation, IWINAC 2022, Proceedings
Editors	José Manuel Ferrández Vicente, José Ramón Álvarez-Sánchez, Félix de la Paz López, Hojjat Adeli
Publisher	Springer Science and Business Media Deutschland GmbH
Pages	3-12
Number of pages	10
ISBN (Print)	9783031062414
DOIs	https://doi.org/10.1007/978-3-031-06242-1_1
Publication status	Published - 2022
Externally published	Yes
Event	9th International Work-Conference on the Interplay Between Natural and Artificial Computation, IWINAC 2022 - Puerto de la Cruz, Spain Duration: 31 May 2022 → 3 Jun 2022

Publication series

Name	Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Volume	13258 LNCS
ISSN (Print)	0302-9743
ISSN (Electronic)	1611-3349

Conference

Conference	9th International Work-Conference on the Interplay Between Natural and Artificial Computation, IWINAC 2022
Country/Territory	Spain
City	Puerto de la Cruz
Period	31/05/22 → 3/06/22

Keywords

Attention mechanism
Deep learning
Medical image

Access to Document

10.1007/978-3-031-06242-1_1

Cite this

Zhu, H., Wang, J., Wang, S. H., Zhang, Y. D., & Górriz, J. M. (2022). ConvNet-CA: A Lightweight Attention-Based CNN for Brain Disease Detection. In J. M. Ferrández Vicente, J. R. Álvarez-Sánchez, F. de la Paz López, & H. Adeli (Eds.), Artificial Intelligence in Neuroscience: Affective Analysis and Health Applications - 9th International Work-Conference on the Interplay Between Natural and Artificial Computation, IWINAC 2022, Proceedings (pp. 3-12). (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 13258 LNCS). Springer Science and Business Media Deutschland GmbH. https://doi.org/10.1007/978-3-031-06242-1_1

Zhu, Hengde ; Wang, Jian ; Wang, Shui Hua et al. / ConvNet-CA : A Lightweight Attention-Based CNN for Brain Disease Detection. Artificial Intelligence in Neuroscience: Affective Analysis and Health Applications - 9th International Work-Conference on the Interplay Between Natural and Artificial Computation, IWINAC 2022, Proceedings. editor / José Manuel Ferrández Vicente ; José Ramón Álvarez-Sánchez ; Félix de la Paz López ; Hojjat Adeli. Springer Science and Business Media Deutschland GmbH, 2022. pp. 3-12 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)).

@inproceedings{6e2c3196d65c47f299b2bd554ed3372f,

title = "ConvNet-CA: A Lightweight Attention-Based CNN for Brain Disease Detection",

abstract = "Attention-based convolutional networks have attracted great interest in recent years and achieved great success in improving representation capability of networks. However, most attention mechanisms are complicated and implemented by introducing a large number of extra parameters. In this study, we proposed a lightweight attention-based convolutional network (ConvNet-CA) that has a low computation complexity yet a high performance for brain disease detection. ConvNet-CA weights the importance of different channels in features maps and pays more attention to important channels by introducing an efficient channel attention mechanism. We evaluated ConvNet-CA on a publicly accessible benchmark dataset: Whole Brain Atlas. The brain diseases involved in this study are stroke, neoplastic disease, degenerative disease, and infectious disease. The experimental results showed that ConvNet-CA achieved highly competitive performance over state-of-the-art methods on distinguishing different types of brain diseases, with an overall multi-class classification accuracy of 94.88 ± 3.64%.",

keywords = "Attention mechanism, Deep learning, Medical image",

author = "Hengde Zhu and Jian Wang and Wang, {Shui Hua} and Zhang, {Yu Dong} and G{\'o}rriz, {Juan M.}",

note = "Publisher Copyright: {\textcopyright} 2022, Springer Nature Switzerland AG.; 9th International Work-Conference on the Interplay Between Natural and Artificial Computation, IWINAC 2022 ; Conference date: 31-05-2022 Through 03-06-2022",

year = "2022",

doi = "10.1007/978-3-031-06242-1_1",

language = "English",

isbn = "9783031062414",

series = "Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)",

publisher = "Springer Science and Business Media Deutschland GmbH",

pages = "3--12",

editor = "{Ferr{\'a}ndez Vicente}, {Jos{\'e} Manuel} and {\'A}lvarez-S{\'a}nchez, {Jos{\'e} Ram{\'o}n} and {de la Paz L{\'o}pez}, F{\'e}lix and Hojjat Adeli",

booktitle = "Artificial Intelligence in Neuroscience",

}

Zhu, H, Wang, J, Wang, SH, Zhang, YD & Górriz, JM 2022, ConvNet-CA: A Lightweight Attention-Based CNN for Brain Disease Detection. in JM Ferrández Vicente, JR Álvarez-Sánchez, F de la Paz López & H Adeli (eds), Artificial Intelligence in Neuroscience: Affective Analysis and Health Applications - 9th International Work-Conference on the Interplay Between Natural and Artificial Computation, IWINAC 2022, Proceedings. Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), vol. 13258 LNCS, Springer Science and Business Media Deutschland GmbH, pp. 3-12, 9th International Work-Conference on the Interplay Between Natural and Artificial Computation, IWINAC 2022, Puerto de la Cruz, Spain, 31/05/22. https://doi.org/10.1007/978-3-031-06242-1_1

ConvNet-CA: A Lightweight Attention-Based CNN for Brain Disease Detection. / Zhu, Hengde; Wang, Jian; Wang, Shui Hua et al.
Artificial Intelligence in Neuroscience: Affective Analysis and Health Applications - 9th International Work-Conference on the Interplay Between Natural and Artificial Computation, IWINAC 2022, Proceedings. ed. / José Manuel Ferrández Vicente; José Ramón Álvarez-Sánchez; Félix de la Paz López; Hojjat Adeli. Springer Science and Business Media Deutschland GmbH, 2022. p. 3-12 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 13258 LNCS).

Research output: Chapter in Book or Report/Conference proceeding › Conference Proceeding › peer-review

TY - GEN

T1 - ConvNet-CA

T2 - 9th International Work-Conference on the Interplay Between Natural and Artificial Computation, IWINAC 2022

AU - Zhu, Hengde

AU - Wang, Jian

AU - Wang, Shui Hua

AU - Zhang, Yu Dong

AU - Górriz, Juan M.

PY - 2022

Y1 - 2022

N2 - Attention-based convolutional networks have attracted great interest in recent years and achieved great success in improving representation capability of networks. However, most attention mechanisms are complicated and implemented by introducing a large number of extra parameters. In this study, we proposed a lightweight attention-based convolutional network (ConvNet-CA) that has a low computation complexity yet a high performance for brain disease detection. ConvNet-CA weights the importance of different channels in features maps and pays more attention to important channels by introducing an efficient channel attention mechanism. We evaluated ConvNet-CA on a publicly accessible benchmark dataset: Whole Brain Atlas. The brain diseases involved in this study are stroke, neoplastic disease, degenerative disease, and infectious disease. The experimental results showed that ConvNet-CA achieved highly competitive performance over state-of-the-art methods on distinguishing different types of brain diseases, with an overall multi-class classification accuracy of 94.88 ± 3.64%.

AB - Attention-based convolutional networks have attracted great interest in recent years and achieved great success in improving representation capability of networks. However, most attention mechanisms are complicated and implemented by introducing a large number of extra parameters. In this study, we proposed a lightweight attention-based convolutional network (ConvNet-CA) that has a low computation complexity yet a high performance for brain disease detection. ConvNet-CA weights the importance of different channels in features maps and pays more attention to important channels by introducing an efficient channel attention mechanism. We evaluated ConvNet-CA on a publicly accessible benchmark dataset: Whole Brain Atlas. The brain diseases involved in this study are stroke, neoplastic disease, degenerative disease, and infectious disease. The experimental results showed that ConvNet-CA achieved highly competitive performance over state-of-the-art methods on distinguishing different types of brain diseases, with an overall multi-class classification accuracy of 94.88 ± 3.64%.

KW - Attention mechanism

KW - Deep learning

KW - Medical image

UR - http://www.scopus.com/inward/record.url?scp=85132014682&partnerID=8YFLogxK

U2 - 10.1007/978-3-031-06242-1_1

DO - 10.1007/978-3-031-06242-1_1

M3 - Conference Proceeding

AN - SCOPUS:85132014682

SN - 9783031062414

T3 - Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)

SP - 3

EP - 12

BT - Artificial Intelligence in Neuroscience

A2 - Ferrández Vicente, José Manuel

A2 - Álvarez-Sánchez, José Ramón

A2 - de la Paz López, Félix

A2 - Adeli, Hojjat

PB - Springer Science and Business Media Deutschland GmbH

Y2 - 31 May 2022 through 3 June 2022

ER -

Zhu H, Wang J, Wang SH, Zhang YD, Górriz JM. ConvNet-CA: A Lightweight Attention-Based CNN for Brain Disease Detection. In Ferrández Vicente JM, Álvarez-Sánchez JR, de la Paz López F, Adeli H, editors, Artificial Intelligence in Neuroscience: Affective Analysis and Health Applications - 9th International Work-Conference on the Interplay Between Natural and Artificial Computation, IWINAC 2022, Proceedings. Springer Science and Business Media Deutschland GmbH. 2022. p. 3-12. (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)). doi: 10.1007/978-3-031-06242-1_1