Leveraging Coupled Interaction for Multimodal Alzheimer's Disease Diagnosis

Yinghuan Shi, Heung Il Suk, Yang Gao, Seong Whan Lee, Dinggang Shen

Research output: Contribution to journalArticle

1 Citation (Scopus)

Abstract

As the population becomes older worldwide, accurate computer-aided diagnosis for Alzheimer's disease (AD) in the early stage has been regarded as a crucial step for neurodegeneration care in recent years. Since it extracts the low-level features from the neuroimaging data, previous methods regarded this computer-aided diagnosis as a classification problem that ignored latent featurewise relation. However, it is known that multiple brain regions in the human brain are anatomically and functionally interlinked according to the current neuroscience perspective. Thus, it is reasonable to assume that the extracted features from different brain regions are related to each other to some extent. Also, the complementary information between different neuroimaging modalities could benefit multimodal fusion. To this end, we consider leveraging the coupled interactions in the feature level and modality level for diagnosis in this paper. First, we propose capturing the feature-level coupled interaction using a coupled feature representation. Then, to model the modality-level coupled interaction, we present two novel methods: 1) the coupled boosting (CB) that models the correlation of pairwise coupled-diversity on both inconsistently and incorrectly classified samples between different modalities and 2) the coupled metric ensemble (CME) that learns an informative feature projection from different modalities by integrating the intrarelation and interrelation of training samples. We systematically evaluated our methods with the AD neuroimaging initiative data set. By comparison with the baseline learning-based methods and the state-of-the-art methods that are specially developed for AD/MCI (mild cognitive impairment) diagnosis, our methods achieved the best performance with accuracy of 95.0% and 80.7% (CB), 94.9% and 79.9% (CME) for AD/NC (normal control), and MCI/NC identification, respectively.

Original languageEnglish
Article number8672088
Pages (from-to)186-200
Number of pages15
JournalIEEE Transactions on Neural Networks and Learning Systems
Volume31
Issue number1
DOIs
Publication statusPublished - 2020 Jan

Fingerprint

Neuroimaging
Alzheimer Disease
Computer aided diagnosis
Brain
Fusion reactions
Neurosciences
Learning
Population

Keywords

  • Computer-aided AD/MCI diagnosis
  • coupled boosting (CB)
  • coupled feature (CFR) representation
  • coupled metric ensemble (CME)

ASJC Scopus subject areas

  • Software
  • Computer Science Applications
  • Computer Networks and Communications
  • Artificial Intelligence

Cite this

Leveraging Coupled Interaction for Multimodal Alzheimer's Disease Diagnosis. / Shi, Yinghuan; Suk, Heung Il; Gao, Yang; Lee, Seong Whan; Shen, Dinggang.

In: IEEE Transactions on Neural Networks and Learning Systems, Vol. 31, No. 1, 8672088, 01.2020, p. 186-200.

Research output: Contribution to journalArticle

@article{8bb17cf25866403e8f1d03ffeafec081,
title = "Leveraging Coupled Interaction for Multimodal Alzheimer's Disease Diagnosis",
abstract = "As the population becomes older worldwide, accurate computer-aided diagnosis for Alzheimer's disease (AD) in the early stage has been regarded as a crucial step for neurodegeneration care in recent years. Since it extracts the low-level features from the neuroimaging data, previous methods regarded this computer-aided diagnosis as a classification problem that ignored latent featurewise relation. However, it is known that multiple brain regions in the human brain are anatomically and functionally interlinked according to the current neuroscience perspective. Thus, it is reasonable to assume that the extracted features from different brain regions are related to each other to some extent. Also, the complementary information between different neuroimaging modalities could benefit multimodal fusion. To this end, we consider leveraging the coupled interactions in the feature level and modality level for diagnosis in this paper. First, we propose capturing the feature-level coupled interaction using a coupled feature representation. Then, to model the modality-level coupled interaction, we present two novel methods: 1) the coupled boosting (CB) that models the correlation of pairwise coupled-diversity on both inconsistently and incorrectly classified samples between different modalities and 2) the coupled metric ensemble (CME) that learns an informative feature projection from different modalities by integrating the intrarelation and interrelation of training samples. We systematically evaluated our methods with the AD neuroimaging initiative data set. By comparison with the baseline learning-based methods and the state-of-the-art methods that are specially developed for AD/MCI (mild cognitive impairment) diagnosis, our methods achieved the best performance with accuracy of 95.0{\%} and 80.7{\%} (CB), 94.9{\%} and 79.9{\%} (CME) for AD/NC (normal control), and MCI/NC identification, respectively.",
keywords = "Computer-aided AD/MCI diagnosis, coupled boosting (CB), coupled feature (CFR) representation, coupled metric ensemble (CME)",
author = "Yinghuan Shi and Suk, {Heung Il} and Yang Gao and Lee, {Seong Whan} and Dinggang Shen",
year = "2020",
month = "1",
doi = "10.1109/TNNLS.2019.2900077",
language = "English",
volume = "31",
pages = "186--200",
journal = "IEEE Transactions on Neural Networks and Learning Systems",
issn = "2162-237X",
publisher = "IEEE Computational Intelligence Society",
number = "1",

}

TY - JOUR

T1 - Leveraging Coupled Interaction for Multimodal Alzheimer's Disease Diagnosis

AU - Shi, Yinghuan

AU - Suk, Heung Il

AU - Gao, Yang

AU - Lee, Seong Whan

AU - Shen, Dinggang

PY - 2020/1

Y1 - 2020/1

N2 - As the population becomes older worldwide, accurate computer-aided diagnosis for Alzheimer's disease (AD) in the early stage has been regarded as a crucial step for neurodegeneration care in recent years. Since it extracts the low-level features from the neuroimaging data, previous methods regarded this computer-aided diagnosis as a classification problem that ignored latent featurewise relation. However, it is known that multiple brain regions in the human brain are anatomically and functionally interlinked according to the current neuroscience perspective. Thus, it is reasonable to assume that the extracted features from different brain regions are related to each other to some extent. Also, the complementary information between different neuroimaging modalities could benefit multimodal fusion. To this end, we consider leveraging the coupled interactions in the feature level and modality level for diagnosis in this paper. First, we propose capturing the feature-level coupled interaction using a coupled feature representation. Then, to model the modality-level coupled interaction, we present two novel methods: 1) the coupled boosting (CB) that models the correlation of pairwise coupled-diversity on both inconsistently and incorrectly classified samples between different modalities and 2) the coupled metric ensemble (CME) that learns an informative feature projection from different modalities by integrating the intrarelation and interrelation of training samples. We systematically evaluated our methods with the AD neuroimaging initiative data set. By comparison with the baseline learning-based methods and the state-of-the-art methods that are specially developed for AD/MCI (mild cognitive impairment) diagnosis, our methods achieved the best performance with accuracy of 95.0% and 80.7% (CB), 94.9% and 79.9% (CME) for AD/NC (normal control), and MCI/NC identification, respectively.

AB - As the population becomes older worldwide, accurate computer-aided diagnosis for Alzheimer's disease (AD) in the early stage has been regarded as a crucial step for neurodegeneration care in recent years. Since it extracts the low-level features from the neuroimaging data, previous methods regarded this computer-aided diagnosis as a classification problem that ignored latent featurewise relation. However, it is known that multiple brain regions in the human brain are anatomically and functionally interlinked according to the current neuroscience perspective. Thus, it is reasonable to assume that the extracted features from different brain regions are related to each other to some extent. Also, the complementary information between different neuroimaging modalities could benefit multimodal fusion. To this end, we consider leveraging the coupled interactions in the feature level and modality level for diagnosis in this paper. First, we propose capturing the feature-level coupled interaction using a coupled feature representation. Then, to model the modality-level coupled interaction, we present two novel methods: 1) the coupled boosting (CB) that models the correlation of pairwise coupled-diversity on both inconsistently and incorrectly classified samples between different modalities and 2) the coupled metric ensemble (CME) that learns an informative feature projection from different modalities by integrating the intrarelation and interrelation of training samples. We systematically evaluated our methods with the AD neuroimaging initiative data set. By comparison with the baseline learning-based methods and the state-of-the-art methods that are specially developed for AD/MCI (mild cognitive impairment) diagnosis, our methods achieved the best performance with accuracy of 95.0% and 80.7% (CB), 94.9% and 79.9% (CME) for AD/NC (normal control), and MCI/NC identification, respectively.

KW - Computer-aided AD/MCI diagnosis

KW - coupled boosting (CB)

KW - coupled feature (CFR) representation

KW - coupled metric ensemble (CME)

UR - http://www.scopus.com/inward/record.url?scp=85068843452&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85068843452&partnerID=8YFLogxK

U2 - 10.1109/TNNLS.2019.2900077

DO - 10.1109/TNNLS.2019.2900077

M3 - Article

C2 - 30908241

AN - SCOPUS:85068843452

VL - 31

SP - 186

EP - 200

JO - IEEE Transactions on Neural Networks and Learning Systems

JF - IEEE Transactions on Neural Networks and Learning Systems

SN - 2162-237X

IS - 1

M1 - 8672088

ER -