A deep residual inception network for HEp-2 cell classification

Yuexiang Li; Linlin Shen

doi:10.1007/978-3-319-67558-9_2

A deep residual inception network for HEp-2 cell classification

Yuexiang Li
, Linlin Shen

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

22 Citations (Scopus)

Abstract

Indirect-immunofluorescence (IIF) of Human Epithelial-2 (HEp-2) cells is a commonly-used method for the diagnosis of autoimmune diseases. Traditional approach relies on specialists to observe HEp-2 slides via the fluorescence microscope, which suffers from a number of shortcomings like being subjective and labor intensive. In this paper, we proposed a hybrid deep learning network combining the latest high-performance network architectures, i.e. ResNet and Inception, to automatically classify HEp-2 cell images. The proposed Deep Residual Inception (DRI) net replaces the plain convolutional layers in Inception with residual modules for better network optimization and fuses the features extracted from shallow, medium and deep layers for performance improvement. The proposed model is evaluated on publicly available I3A (Indirect Immunofluorescence Image Analysis) dataset. The experiment results demonstrate that our proposed DRI remarkably outperforms the benchmarking approaches.

Original language	English
Title of host publication	Deep Learning in Medical Image Analysis and Multimodal Learning for Clinical Decision Support - 3rd International Workshop, DLMIA 2017 and 7th International Workshop, ML-CDS 2017 Held in Conjunction with MICCAI 2017, Proceedings
Editors	Tal Arbel, M. Jorge Cardoso
Publisher	Springer Verlag
Pages	12-20
Number of pages	9
ISBN (Print)	9783319675572
DOIs	https://doi.org/10.1007/978-3-319-67558-9_2
Publication status	Published - 2017
Externally published	Yes
Event	3rd International Workshop on Deep Learning in Medical Image Analysis, DLMIA 2017 and 7th International Workshop on Multimodal Learning for Clinical Decision Support, ML-CDS 2017 held in Conjunction with 20th International Conference on Medical Image Computing and Computer Assisted Intervention, MICCAI 2017 - Quebec City, Canada Duration: 14 Sept 2017 → 14 Sept 2017

Publication series

Name	Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Volume	10553 LNCS
ISSN (Print)	0302-9743
ISSN (Electronic)	1611-3349

Conference

Conference	3rd International Workshop on Deep Learning in Medical Image Analysis, DLMIA 2017 and 7th International Workshop on Multimodal Learning for Clinical Decision Support, ML-CDS 2017 held in Conjunction with 20th International Conference on Medical Image Computing and Computer Assisted Intervention, MICCAI 2017
Country/Territory	Canada
City	Quebec City
Period	14/09/17 → 14/09/17

UN SDGs

This output contributes to the following UN Sustainable Development Goals (SDGs)

SDG 3 Good Health and Well-being

Free Keywords

Deep learning network
HEp-2 cells
Image classification

ASJC Scopus subject areas

Theoretical Computer Science
General Computer Science

Access to Document

10.1007/978-3-319-67558-9_2

Cite this

Li, Y., & Shen, L. (2017). A deep residual inception network for HEp-2 cell classification. In T. Arbel, & M. J. Cardoso (Eds.), Deep Learning in Medical Image Analysis and Multimodal Learning for Clinical Decision Support - 3rd International Workshop, DLMIA 2017 and 7th International Workshop, ML-CDS 2017 Held in Conjunction with MICCAI 2017, Proceedings (pp. 12-20). (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 10553 LNCS). Springer Verlag. https://doi.org/10.1007/978-3-319-67558-9_2

Li, Yuexiang ; Shen, Linlin. / A deep residual inception network for HEp-2 cell classification. Deep Learning in Medical Image Analysis and Multimodal Learning for Clinical Decision Support - 3rd International Workshop, DLMIA 2017 and 7th International Workshop, ML-CDS 2017 Held in Conjunction with MICCAI 2017, Proceedings. editor / Tal Arbel ; M. Jorge Cardoso. Springer Verlag, 2017. pp. 12-20 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)).

@inproceedings{7f743d5ea12f46c5a14b94d3cfaaa786,

title = "A deep residual inception network for HEp-2 cell classification",

abstract = "Indirect-immunofluorescence (IIF) of Human Epithelial-2 (HEp-2) cells is a commonly-used method for the diagnosis of autoimmune diseases. Traditional approach relies on specialists to observe HEp-2 slides via the fluorescence microscope, which suffers from a number of shortcomings like being subjective and labor intensive. In this paper, we proposed a hybrid deep learning network combining the latest high-performance network architectures, i.e. ResNet and Inception, to automatically classify HEp-2 cell images. The proposed Deep Residual Inception (DRI) net replaces the plain convolutional layers in Inception with residual modules for better network optimization and fuses the features extracted from shallow, medium and deep layers for performance improvement. The proposed model is evaluated on publicly available I3A (Indirect Immunofluorescence Image Analysis) dataset. The experiment results demonstrate that our proposed DRI remarkably outperforms the benchmarking approaches.",

keywords = "Deep learning network, HEp-2 cells, Image classification",

author = "Yuexiang Li and Linlin Shen",

note = "Publisher Copyright: {\textcopyright} Springer International Publishing AG 2017.; 3rd International Workshop on Deep Learning in Medical Image Analysis, DLMIA 2017 and 7th International Workshop on Multimodal Learning for Clinical Decision Support, ML-CDS 2017 held in Conjunction with 20th International Conference on Medical Image Computing and Computer Assisted Intervention, MICCAI 2017 ; Conference date: 14-09-2017 Through 14-09-2017",

year = "2017",

doi = "10.1007/978-3-319-67558-9\_2",

language = "English",

isbn = "9783319675572",

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

publisher = "Springer Verlag",

pages = "12--20",

editor = "Tal Arbel and Cardoso, \{M. Jorge\}",

booktitle = "Deep Learning in Medical Image Analysis and Multimodal Learning for Clinical Decision Support - 3rd International Workshop, DLMIA 2017 and 7th International Workshop, ML-CDS 2017 Held in Conjunction with MICCAI 2017, Proceedings",

address = "Germany",

}

Li, Y & Shen, L 2017, A deep residual inception network for HEp-2 cell classification. in T Arbel & MJ Cardoso (eds), Deep Learning in Medical Image Analysis and Multimodal Learning for Clinical Decision Support - 3rd International Workshop, DLMIA 2017 and 7th International Workshop, ML-CDS 2017 Held in Conjunction with MICCAI 2017, Proceedings. Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), vol. 10553 LNCS, Springer Verlag, pp. 12-20, 3rd International Workshop on Deep Learning in Medical Image Analysis, DLMIA 2017 and 7th International Workshop on Multimodal Learning for Clinical Decision Support, ML-CDS 2017 held in Conjunction with 20th International Conference on Medical Image Computing and Computer Assisted Intervention, MICCAI 2017, Quebec City, Canada, 14/09/17. https://doi.org/10.1007/978-3-319-67558-9_2

A deep residual inception network for HEp-2 cell classification. / Li, Yuexiang; Shen, Linlin.
Deep Learning in Medical Image Analysis and Multimodal Learning for Clinical Decision Support - 3rd International Workshop, DLMIA 2017 and 7th International Workshop, ML-CDS 2017 Held in Conjunction with MICCAI 2017, Proceedings. ed. / Tal Arbel; M. Jorge Cardoso. Springer Verlag, 2017. p. 12-20 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 10553 LNCS).

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

TY - GEN

T1 - A deep residual inception network for HEp-2 cell classification

AU - Li, Yuexiang

AU - Shen, Linlin

N1 - Publisher Copyright: © Springer International Publishing AG 2017.

PY - 2017

Y1 - 2017

N2 - Indirect-immunofluorescence (IIF) of Human Epithelial-2 (HEp-2) cells is a commonly-used method for the diagnosis of autoimmune diseases. Traditional approach relies on specialists to observe HEp-2 slides via the fluorescence microscope, which suffers from a number of shortcomings like being subjective and labor intensive. In this paper, we proposed a hybrid deep learning network combining the latest high-performance network architectures, i.e. ResNet and Inception, to automatically classify HEp-2 cell images. The proposed Deep Residual Inception (DRI) net replaces the plain convolutional layers in Inception with residual modules for better network optimization and fuses the features extracted from shallow, medium and deep layers for performance improvement. The proposed model is evaluated on publicly available I3A (Indirect Immunofluorescence Image Analysis) dataset. The experiment results demonstrate that our proposed DRI remarkably outperforms the benchmarking approaches.

AB - Indirect-immunofluorescence (IIF) of Human Epithelial-2 (HEp-2) cells is a commonly-used method for the diagnosis of autoimmune diseases. Traditional approach relies on specialists to observe HEp-2 slides via the fluorescence microscope, which suffers from a number of shortcomings like being subjective and labor intensive. In this paper, we proposed a hybrid deep learning network combining the latest high-performance network architectures, i.e. ResNet and Inception, to automatically classify HEp-2 cell images. The proposed Deep Residual Inception (DRI) net replaces the plain convolutional layers in Inception with residual modules for better network optimization and fuses the features extracted from shallow, medium and deep layers for performance improvement. The proposed model is evaluated on publicly available I3A (Indirect Immunofluorescence Image Analysis) dataset. The experiment results demonstrate that our proposed DRI remarkably outperforms the benchmarking approaches.

KW - Deep learning network

KW - HEp-2 cells

KW - Image classification

UR - https://www.scopus.com/pages/publications/85029788200

U2 - 10.1007/978-3-319-67558-9_2

DO - 10.1007/978-3-319-67558-9_2

M3 - Conference contribution

AN - SCOPUS:85029788200

SN - 9783319675572

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

SP - 12

EP - 20

BT - Deep Learning in Medical Image Analysis and Multimodal Learning for Clinical Decision Support - 3rd International Workshop, DLMIA 2017 and 7th International Workshop, ML-CDS 2017 Held in Conjunction with MICCAI 2017, Proceedings

A2 - Arbel, Tal

A2 - Cardoso, M. Jorge

PB - Springer Verlag

T2 - 3rd International Workshop on Deep Learning in Medical Image Analysis, DLMIA 2017 and 7th International Workshop on Multimodal Learning for Clinical Decision Support, ML-CDS 2017 held in Conjunction with 20th International Conference on Medical Image Computing and Computer Assisted Intervention, MICCAI 2017

Y2 - 14 September 2017 through 14 September 2017

ER -

Li Y, Shen L. A deep residual inception network for HEp-2 cell classification. In Arbel T, Cardoso MJ, editors, Deep Learning in Medical Image Analysis and Multimodal Learning for Clinical Decision Support - 3rd International Workshop, DLMIA 2017 and 7th International Workshop, ML-CDS 2017 Held in Conjunction with MICCAI 2017, Proceedings. Springer Verlag. 2017. p. 12-20. (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)). doi: 10.1007/978-3-319-67558-9_2

A deep residual inception network for HEp-2 cell classification

Abstract

Publication series

Conference

UN SDGs

Free Keywords

ASJC Scopus subject areas

Access to Document

Other files and links

Fingerprint

Cite this