Co-neighbor multi-view spectral embedding for medical content-based retrieval

Hangyu Che; Sidong Liu; Weidong Cai; Sonia Pujol; Ron Kikinis; Dagan Feng

doi:10.1109/ISBI.2014.6868019

Co-neighbor multi-view spectral embedding for medical content-based retrieval

Hangyu Che, Sidong Liu, Weidong Cai, Sonia Pujol, Ron Kikinis, Dagan Feng

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

10 Citations (Scopus)

Abstract

Multimodal medical data from various information sources are often used to depict patients. We refer to each source as a 'view'. Multi-view features could provide complementary information to each other; thus by fusing the multi-view features, we could greatly enhance the current medical content-based retrieval framework. In this paper, we propose a Co-neighbor Multi-view Spectral Embedding (CMSE) algorithm, which is an advanced feature fusion method based on the multi-view spectral analysis. CMSE aims to seek a smooth embedding for the multi-view features by maximizing the neighborhood affinity across all feature spaces. We evaluated the proposed CMSE algorithm using a freely available neuroimaging database, ADNI, with 331 pre-diagnosed subjects. Totally, 9 views of features were extracted for validation, and an improved retrieval performance was achieved over other state-of-art feature fusion methods.

Original language	English
Title of host publication	2014 IEEE 11th International Symposium on Biomedical Imaging, ISBI 2014
Publisher	Institute of Electrical and Electronics Engineers (IEEE)
Pages	911-914
Number of pages	4
ISBN (Electronic)	9781467319591, 9781467319614
DOIs	https://doi.org/10.1109/ISBI.2014.6868019
Publication status	Published - 1 Jan 2014
Externally published	Yes
Event	2014 IEEE 11th International Symposium on Biomedical Imaging, ISBI 2014 - Beijing, China Duration: 29 Apr 2014 → 2 May 2014

Conference

Conference	2014 IEEE 11th International Symposium on Biomedical Imaging, ISBI 2014
Country/Territory	China
City	Beijing
Period	29/04/14 → 2/05/14

Keywords

Dimensionality reduction
Multiple views
Neuroimaging
Spectral embedding

Access to Document

10.1109/ISBI.2014.6868019

Cite this

@inproceedings{b7fb76f0bee9466c8395aed33aa9e06c,

title = "Co-neighbor multi-view spectral embedding for medical content-based retrieval",

abstract = "Multimodal medical data from various information sources are often used to depict patients. We refer to each source as a 'view'. Multi-view features could provide complementary information to each other; thus by fusing the multi-view features, we could greatly enhance the current medical content-based retrieval framework. In this paper, we propose a Co-neighbor Multi-view Spectral Embedding (CMSE) algorithm, which is an advanced feature fusion method based on the multi-view spectral analysis. CMSE aims to seek a smooth embedding for the multi-view features by maximizing the neighborhood affinity across all feature spaces. We evaluated the proposed CMSE algorithm using a freely available neuroimaging database, ADNI, with 331 pre-diagnosed subjects. Totally, 9 views of features were extracted for validation, and an improved retrieval performance was achieved over other state-of-art feature fusion methods.",

keywords = "Dimensionality reduction, Multiple views, Neuroimaging, Spectral embedding",

author = "Hangyu Che and Sidong Liu and Weidong Cai and Sonia Pujol and Ron Kikinis and Dagan Feng",

year = "2014",

month = jan,

day = "1",

doi = "10.1109/ISBI.2014.6868019",

language = "English",

pages = "911--914",

booktitle = "2014 IEEE 11th International Symposium on Biomedical Imaging, ISBI 2014",

publisher = "Institute of Electrical and Electronics Engineers (IEEE)",

address = "United States",

note = "2014 IEEE 11th International Symposium on Biomedical Imaging, ISBI 2014 ; Conference date: 29-04-2014 Through 02-05-2014",

}

Che, H, Liu, S, Cai, W, Pujol, S, Kikinis, R & Feng, D 2014, Co-neighbor multi-view spectral embedding for medical content-based retrieval. in 2014 IEEE 11th International Symposium on Biomedical Imaging, ISBI 2014., 6868019, Institute of Electrical and Electronics Engineers (IEEE), pp. 911-914, 2014 IEEE 11th International Symposium on Biomedical Imaging, ISBI 2014, Beijing, China, 29/04/14. https://doi.org/10.1109/ISBI.2014.6868019

Co-neighbor multi-view spectral embedding for medical content-based retrieval. / Che, Hangyu; Liu, Sidong; Cai, Weidong et al.
2014 IEEE 11th International Symposium on Biomedical Imaging, ISBI 2014. Institute of Electrical and Electronics Engineers (IEEE), 2014. p. 911-914 6868019.

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

TY - GEN

T1 - Co-neighbor multi-view spectral embedding for medical content-based retrieval

AU - Che, Hangyu

AU - Liu, Sidong

AU - Cai, Weidong

AU - Pujol, Sonia

AU - Kikinis, Ron

AU - Feng, Dagan

PY - 2014/1/1

Y1 - 2014/1/1

N2 - Multimodal medical data from various information sources are often used to depict patients. We refer to each source as a 'view'. Multi-view features could provide complementary information to each other; thus by fusing the multi-view features, we could greatly enhance the current medical content-based retrieval framework. In this paper, we propose a Co-neighbor Multi-view Spectral Embedding (CMSE) algorithm, which is an advanced feature fusion method based on the multi-view spectral analysis. CMSE aims to seek a smooth embedding for the multi-view features by maximizing the neighborhood affinity across all feature spaces. We evaluated the proposed CMSE algorithm using a freely available neuroimaging database, ADNI, with 331 pre-diagnosed subjects. Totally, 9 views of features were extracted for validation, and an improved retrieval performance was achieved over other state-of-art feature fusion methods.

AB - Multimodal medical data from various information sources are often used to depict patients. We refer to each source as a 'view'. Multi-view features could provide complementary information to each other; thus by fusing the multi-view features, we could greatly enhance the current medical content-based retrieval framework. In this paper, we propose a Co-neighbor Multi-view Spectral Embedding (CMSE) algorithm, which is an advanced feature fusion method based on the multi-view spectral analysis. CMSE aims to seek a smooth embedding for the multi-view features by maximizing the neighborhood affinity across all feature spaces. We evaluated the proposed CMSE algorithm using a freely available neuroimaging database, ADNI, with 331 pre-diagnosed subjects. Totally, 9 views of features were extracted for validation, and an improved retrieval performance was achieved over other state-of-art feature fusion methods.

KW - Dimensionality reduction

KW - Multiple views

KW - Neuroimaging

KW - Spectral embedding

UR - https://www.scopus.com/pages/publications/84920890201

U2 - 10.1109/ISBI.2014.6868019

DO - 10.1109/ISBI.2014.6868019

M3 - Conference proceeding contribution

AN - SCOPUS:84920890201

SP - 911

EP - 914

BT - 2014 IEEE 11th International Symposium on Biomedical Imaging, ISBI 2014

PB - Institute of Electrical and Electronics Engineers (IEEE)

T2 - 2014 IEEE 11th International Symposium on Biomedical Imaging, ISBI 2014

Y2 - 29 April 2014 through 2 May 2014

ER -

Co-neighbor multi-view spectral embedding for medical content-based retrieval

Abstract

Conference

Keywords

Access to Document

Other files and links

Fingerprint

Cite this