A Deep Concatenated Convolutional Neural Network-Based Method to Classify Autism

Tanu Wadhera; Mufti Mahmud; David J. Brown

doi:10.1007/978-981-99-1648-1_37

A Deep Concatenated Convolutional Neural Network-Based Method to Classify Autism

Tanu Wadhera
, Mufti Mahmud^*
, David J. Brown

^*Corresponding author for this work

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

9 Scopus citations

Abstract

Associating different brain regions relating to a particular neurological issue has emerged as an area of neuroimaging research. Deep learning algorithms have emerged as a promising approach to automate neural data processing for classifying traits or characteristics associated with a range of conditions. The present paper has worked on improving binary-classification accuracy of Autism Spectrum Disorder (ASD) by distinguishing ASD from Typically Developing (TD) individuals. A hybrid model is proposed concatenating VGGNet and ResNet-152 to fuse the most discerning heterogeneous features from both networks to build a strong feature vector for attaining high classification accuracy. The effectiveness of the proposed approach is demonstrated on ABIDE dataset, which showed an improvement over state-of-art classifiers in terms of accuracy (88.12%), sensitivity (91.32%), specificity (86.34%) and ROC (0.88), respectively, in classifying ASD and TD individuals.

Original language	English
Title of host publication	Neural Information Processing - 29th International Conference, ICONIP 2022, Proceedings
Editors	Mohammad Tanveer, Sonali Agarwal, Seiichi Ozawa, Asif Ekbal, Adam Jatowt
Publisher	Springer Science and Business Media Deutschland GmbH
Pages	446-458
Number of pages	13
ISBN (Print)	9789819916474
DOIs	https://doi.org/10.1007/978-981-99-1648-1_37
State	Published - 2023
Externally published	Yes
Event	29th International Conference on Neural Information Processing, ICONIP 2022 - Virtual, Online Duration: 22 Nov 2022 → 26 Nov 2022

Publication series

Name	Communications in Computer and Information Science
Volume	1794 CCIS
ISSN (Print)	1865-0929
ISSN (Electronic)	1865-0937

Conference

Conference	29th International Conference on Neural Information Processing, ICONIP 2022
City	Virtual, Online
Period	22/11/22 → 26/11/22

Bibliographical note

Publisher Copyright:
© 2023, The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd.

Keywords

ABIDE
Autism
Deep Learning
Fusion
ResNet-52
VGGNet

ASJC Scopus subject areas

General Computer Science
General Mathematics

Access to Document

10.1007/978-981-99-1648-1_37

Cite this

Wadhera, T., Mahmud, M., & Brown, D. J. (2023). A Deep Concatenated Convolutional Neural Network-Based Method to Classify Autism. In M. Tanveer, S. Agarwal, S. Ozawa, A. Ekbal, & A. Jatowt (Eds.), Neural Information Processing - 29th International Conference, ICONIP 2022, Proceedings (pp. 446-458). (Communications in Computer and Information Science; Vol. 1794 CCIS). Springer Science and Business Media Deutschland GmbH. https://doi.org/10.1007/978-981-99-1648-1_37

Wadhera, Tanu ; Mahmud, Mufti ; Brown, David J. / A Deep Concatenated Convolutional Neural Network-Based Method to Classify Autism. Neural Information Processing - 29th International Conference, ICONIP 2022, Proceedings. editor / Mohammad Tanveer ; Sonali Agarwal ; Seiichi Ozawa ; Asif Ekbal ; Adam Jatowt. Springer Science and Business Media Deutschland GmbH, 2023. pp. 446-458 (Communications in Computer and Information Science).

@inproceedings{dc6e535d08ed4021b93b7859bd74b83d,

title = "A Deep Concatenated Convolutional Neural Network-Based Method to Classify Autism",

abstract = "Associating different brain regions relating to a particular neurological issue has emerged as an area of neuroimaging research. Deep learning algorithms have emerged as a promising approach to automate neural data processing for classifying traits or characteristics associated with a range of conditions. The present paper has worked on improving binary-classification accuracy of Autism Spectrum Disorder (ASD) by distinguishing ASD from Typically Developing (TD) individuals. A hybrid model is proposed concatenating VGGNet and ResNet-152 to fuse the most discerning heterogeneous features from both networks to build a strong feature vector for attaining high classification accuracy. The effectiveness of the proposed approach is demonstrated on ABIDE dataset, which showed an improvement over state-of-art classifiers in terms of accuracy (88.12\%), sensitivity (91.32\%), specificity (86.34\%) and ROC (0.88), respectively, in classifying ASD and TD individuals.",

keywords = "ABIDE, Autism, Deep Learning, Fusion, ResNet-52, VGGNet",

author = "Tanu Wadhera and Mufti Mahmud and Brown, \{David J.\}",

note = "Publisher Copyright: {\textcopyright} 2023, The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd.; 29th International Conference on Neural Information Processing, ICONIP 2022 ; Conference date: 22-11-2022 Through 26-11-2022",

year = "2023",

doi = "10.1007/978-981-99-1648-1\_37",

language = "English",

isbn = "9789819916474",

series = "Communications in Computer and Information Science",

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

pages = "446--458",

editor = "Mohammad Tanveer and Sonali Agarwal and Seiichi Ozawa and Asif Ekbal and Adam Jatowt",

booktitle = "Neural Information Processing - 29th International Conference, ICONIP 2022, Proceedings",

address = "Germany",

}

Wadhera, T, Mahmud, M & Brown, DJ 2023, A Deep Concatenated Convolutional Neural Network-Based Method to Classify Autism. in M Tanveer, S Agarwal, S Ozawa, A Ekbal & A Jatowt (eds), Neural Information Processing - 29th International Conference, ICONIP 2022, Proceedings. Communications in Computer and Information Science, vol. 1794 CCIS, Springer Science and Business Media Deutschland GmbH, pp. 446-458, 29th International Conference on Neural Information Processing, ICONIP 2022, Virtual, Online, 22/11/22. https://doi.org/10.1007/978-981-99-1648-1_37

A Deep Concatenated Convolutional Neural Network-Based Method to Classify Autism. / Wadhera, Tanu; Mahmud, Mufti; Brown, David J.
Neural Information Processing - 29th International Conference, ICONIP 2022, Proceedings. ed. / Mohammad Tanveer; Sonali Agarwal; Seiichi Ozawa; Asif Ekbal; Adam Jatowt. Springer Science and Business Media Deutschland GmbH, 2023. p. 446-458 (Communications in Computer and Information Science; Vol. 1794 CCIS).

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

TY - GEN

T1 - A Deep Concatenated Convolutional Neural Network-Based Method to Classify Autism

AU - Wadhera, Tanu

AU - Mahmud, Mufti

AU - Brown, David J.

PY - 2023

Y1 - 2023

N2 - Associating different brain regions relating to a particular neurological issue has emerged as an area of neuroimaging research. Deep learning algorithms have emerged as a promising approach to automate neural data processing for classifying traits or characteristics associated with a range of conditions. The present paper has worked on improving binary-classification accuracy of Autism Spectrum Disorder (ASD) by distinguishing ASD from Typically Developing (TD) individuals. A hybrid model is proposed concatenating VGGNet and ResNet-152 to fuse the most discerning heterogeneous features from both networks to build a strong feature vector for attaining high classification accuracy. The effectiveness of the proposed approach is demonstrated on ABIDE dataset, which showed an improvement over state-of-art classifiers in terms of accuracy (88.12%), sensitivity (91.32%), specificity (86.34%) and ROC (0.88), respectively, in classifying ASD and TD individuals.

AB - Associating different brain regions relating to a particular neurological issue has emerged as an area of neuroimaging research. Deep learning algorithms have emerged as a promising approach to automate neural data processing for classifying traits or characteristics associated with a range of conditions. The present paper has worked on improving binary-classification accuracy of Autism Spectrum Disorder (ASD) by distinguishing ASD from Typically Developing (TD) individuals. A hybrid model is proposed concatenating VGGNet and ResNet-152 to fuse the most discerning heterogeneous features from both networks to build a strong feature vector for attaining high classification accuracy. The effectiveness of the proposed approach is demonstrated on ABIDE dataset, which showed an improvement over state-of-art classifiers in terms of accuracy (88.12%), sensitivity (91.32%), specificity (86.34%) and ROC (0.88), respectively, in classifying ASD and TD individuals.

KW - ABIDE

KW - Autism

KW - Deep Learning

KW - Fusion

KW - ResNet-52

KW - VGGNet

UR - https://www.scopus.com/pages/publications/85161723191

U2 - 10.1007/978-981-99-1648-1_37

DO - 10.1007/978-981-99-1648-1_37

M3 - Conference contribution

AN - SCOPUS:85161723191

SN - 9789819916474

T3 - Communications in Computer and Information Science

SP - 446

EP - 458

BT - Neural Information Processing - 29th International Conference, ICONIP 2022, Proceedings

A2 - Tanveer, Mohammad

A2 - Agarwal, Sonali

A2 - Ozawa, Seiichi

A2 - Ekbal, Asif

A2 - Jatowt, Adam

PB - Springer Science and Business Media Deutschland GmbH

T2 - 29th International Conference on Neural Information Processing, ICONIP 2022

Y2 - 22 November 2022 through 26 November 2022

ER -

Wadhera T, Mahmud M, Brown DJ. A Deep Concatenated Convolutional Neural Network-Based Method to Classify Autism. In Tanveer M, Agarwal S, Ozawa S, Ekbal A, Jatowt A, editors, Neural Information Processing - 29th International Conference, ICONIP 2022, Proceedings. Springer Science and Business Media Deutschland GmbH. 2023. p. 446-458. (Communications in Computer and Information Science). doi: 10.1007/978-981-99-1648-1_37