Relationship induced multi-atlas learning for Alzheimer’s disease diagnosis

Mingxia Liu, Daoqiang Zhang, Ehsan Adeli-Mosabbeb, Dinggang Shen

Research output: Chapter in Book/Report/Conference proceedingConference contribution

1 Citation (Scopus)

Abstract

Multi-atlas based methods using magnetic resonance imaging (MRI) have been recently proposed for automatic diagnosis of Alzheimer’s disease (AD) and its prodromal stage, i.e., mild cognitive impairment (MCI). However, most existing multi-atlas based methods simply average or concatenate features generated from multiple atlases, which ignores the important underlying structure information of multiatlas data. In this paper, we propose a novel relationship induced multiatlas learning (RIML) method for AD/MCI classification. Specifically, we first register each brain image onto multiple selected atlases separately, through which multiple sets of feature representations can be extracted. To exploit the structure information of data, we develop a relationship induced sparse feature selection method, by employing two regularization terms to model the relationships among atlases and among subjects. Finally, we learn a classifier based on selected features in each atlas space, followed by an ensemble classification strategy to combine multiple classifiers for making a final decision. Experimental results on the Alzheimer’s Disease Neuroimaging Initiative (ADNI) database demonstrate that our method achieves significant performance improvement for AD/MCI classification, compared with several state-of-the-art methods.

Original languageEnglish
Title of host publicationMedical Computer Vision: Algorithms for Big Data - International Workshop, MCV 2015 and Held in Conjunction with MICCAI 2015, Revised Selected Papers
PublisherSpringer Verlag
Pages24-33
Number of pages10
Volume9601
ISBN (Print)9783319420158
DOIs
Publication statusPublished - 2016
Externally publishedYes
EventInternational Workshop on Medical Image Computing for Computer Assisted Intervention, 2015 MICCAI - Germany, Germany
Duration: 2015 Oct 92015 Oct 9

Publication series

NameLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Volume9601
ISSN (Print)03029743
ISSN (Electronic)16113349

Other

OtherInternational Workshop on Medical Image Computing for Computer Assisted Intervention, 2015 MICCAI
CountryGermany
CityGermany
Period15/10/915/10/9

Fingerprint

Alzheimer's Disease
Atlas
Classifiers
Information Structure
Neuroimaging
Magnetic resonance
Multiple Classifiers
Feature extraction
Brain
Magnetic Resonance Imaging
Imaging techniques
Feature Selection
Learning
Relationships
Regularization
Ensemble
Classifier
Experimental Results
Term
Demonstrate

ASJC Scopus subject areas

  • Theoretical Computer Science
  • Computer Science(all)

Cite this

Liu, M., Zhang, D., Adeli-Mosabbeb, E., & Shen, D. (2016). Relationship induced multi-atlas learning for Alzheimer’s disease diagnosis. In Medical Computer Vision: Algorithms for Big Data - International Workshop, MCV 2015 and Held in Conjunction with MICCAI 2015, Revised Selected Papers (Vol. 9601, pp. 24-33). (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 9601). Springer Verlag. https://doi.org/10.1007/978-3-319-42016-5_3

Relationship induced multi-atlas learning for Alzheimer’s disease diagnosis. / Liu, Mingxia; Zhang, Daoqiang; Adeli-Mosabbeb, Ehsan; Shen, Dinggang.

Medical Computer Vision: Algorithms for Big Data - International Workshop, MCV 2015 and Held in Conjunction with MICCAI 2015, Revised Selected Papers. Vol. 9601 Springer Verlag, 2016. p. 24-33 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 9601).

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Liu, M, Zhang, D, Adeli-Mosabbeb, E & Shen, D 2016, Relationship induced multi-atlas learning for Alzheimer’s disease diagnosis. in Medical Computer Vision: Algorithms for Big Data - International Workshop, MCV 2015 and Held in Conjunction with MICCAI 2015, Revised Selected Papers. vol. 9601, Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), vol. 9601, Springer Verlag, pp. 24-33, International Workshop on Medical Image Computing for Computer Assisted Intervention, 2015 MICCAI, Germany, Germany, 15/10/9. https://doi.org/10.1007/978-3-319-42016-5_3
Liu M, Zhang D, Adeli-Mosabbeb E, Shen D. Relationship induced multi-atlas learning for Alzheimer’s disease diagnosis. In Medical Computer Vision: Algorithms for Big Data - International Workshop, MCV 2015 and Held in Conjunction with MICCAI 2015, Revised Selected Papers. Vol. 9601. Springer Verlag. 2016. p. 24-33. (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)). https://doi.org/10.1007/978-3-319-42016-5_3
Liu, Mingxia ; Zhang, Daoqiang ; Adeli-Mosabbeb, Ehsan ; Shen, Dinggang. / Relationship induced multi-atlas learning for Alzheimer’s disease diagnosis. Medical Computer Vision: Algorithms for Big Data - International Workshop, MCV 2015 and Held in Conjunction with MICCAI 2015, Revised Selected Papers. Vol. 9601 Springer Verlag, 2016. pp. 24-33 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)).
@inproceedings{94e56f5279f24ae69007585892714994,
title = "Relationship induced multi-atlas learning for Alzheimer’s disease diagnosis",
abstract = "Multi-atlas based methods using magnetic resonance imaging (MRI) have been recently proposed for automatic diagnosis of Alzheimer’s disease (AD) and its prodromal stage, i.e., mild cognitive impairment (MCI). However, most existing multi-atlas based methods simply average or concatenate features generated from multiple atlases, which ignores the important underlying structure information of multiatlas data. In this paper, we propose a novel relationship induced multiatlas learning (RIML) method for AD/MCI classification. Specifically, we first register each brain image onto multiple selected atlases separately, through which multiple sets of feature representations can be extracted. To exploit the structure information of data, we develop a relationship induced sparse feature selection method, by employing two regularization terms to model the relationships among atlases and among subjects. Finally, we learn a classifier based on selected features in each atlas space, followed by an ensemble classification strategy to combine multiple classifiers for making a final decision. Experimental results on the Alzheimer’s Disease Neuroimaging Initiative (ADNI) database demonstrate that our method achieves significant performance improvement for AD/MCI classification, compared with several state-of-the-art methods.",
author = "Mingxia Liu and Daoqiang Zhang and Ehsan Adeli-Mosabbeb and Dinggang Shen",
year = "2016",
doi = "10.1007/978-3-319-42016-5_3",
language = "English",
isbn = "9783319420158",
volume = "9601",
series = "Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)",
publisher = "Springer Verlag",
pages = "24--33",
booktitle = "Medical Computer Vision: Algorithms for Big Data - International Workshop, MCV 2015 and Held in Conjunction with MICCAI 2015, Revised Selected Papers",

}

TY - GEN

T1 - Relationship induced multi-atlas learning for Alzheimer’s disease diagnosis

AU - Liu, Mingxia

AU - Zhang, Daoqiang

AU - Adeli-Mosabbeb, Ehsan

AU - Shen, Dinggang

PY - 2016

Y1 - 2016

N2 - Multi-atlas based methods using magnetic resonance imaging (MRI) have been recently proposed for automatic diagnosis of Alzheimer’s disease (AD) and its prodromal stage, i.e., mild cognitive impairment (MCI). However, most existing multi-atlas based methods simply average or concatenate features generated from multiple atlases, which ignores the important underlying structure information of multiatlas data. In this paper, we propose a novel relationship induced multiatlas learning (RIML) method for AD/MCI classification. Specifically, we first register each brain image onto multiple selected atlases separately, through which multiple sets of feature representations can be extracted. To exploit the structure information of data, we develop a relationship induced sparse feature selection method, by employing two regularization terms to model the relationships among atlases and among subjects. Finally, we learn a classifier based on selected features in each atlas space, followed by an ensemble classification strategy to combine multiple classifiers for making a final decision. Experimental results on the Alzheimer’s Disease Neuroimaging Initiative (ADNI) database demonstrate that our method achieves significant performance improvement for AD/MCI classification, compared with several state-of-the-art methods.

AB - Multi-atlas based methods using magnetic resonance imaging (MRI) have been recently proposed for automatic diagnosis of Alzheimer’s disease (AD) and its prodromal stage, i.e., mild cognitive impairment (MCI). However, most existing multi-atlas based methods simply average or concatenate features generated from multiple atlases, which ignores the important underlying structure information of multiatlas data. In this paper, we propose a novel relationship induced multiatlas learning (RIML) method for AD/MCI classification. Specifically, we first register each brain image onto multiple selected atlases separately, through which multiple sets of feature representations can be extracted. To exploit the structure information of data, we develop a relationship induced sparse feature selection method, by employing two regularization terms to model the relationships among atlases and among subjects. Finally, we learn a classifier based on selected features in each atlas space, followed by an ensemble classification strategy to combine multiple classifiers for making a final decision. Experimental results on the Alzheimer’s Disease Neuroimaging Initiative (ADNI) database demonstrate that our method achieves significant performance improvement for AD/MCI classification, compared with several state-of-the-art methods.

UR - http://www.scopus.com/inward/record.url?scp=84981303427&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84981303427&partnerID=8YFLogxK

U2 - 10.1007/978-3-319-42016-5_3

DO - 10.1007/978-3-319-42016-5_3

M3 - Conference contribution

AN - SCOPUS:84981303427

SN - 9783319420158

VL - 9601

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

SP - 24

EP - 33

BT - Medical Computer Vision: Algorithms for Big Data - International Workshop, MCV 2015 and Held in Conjunction with MICCAI 2015, Revised Selected Papers

PB - Springer Verlag

ER -