Sparse Multiview Task-Centralized Ensemble Learning for ASD Diagnosis Based on Age- and Sex-Related Functional Connectivity Patterns

Jun Wang, Qian Wang, Han Zhang, Jiawei Chen, Shitong Wang, Dinggang Shen

Research output: Contribution to journalArticle

6 Citations (Scopus)

Abstract

Autism spectrum disorder (ASD) is an age- and sex-related neurodevelopmental disorder that alters the brain's functional connectivity (FC). The changes caused by ASD are associated with different age- and sex-related patterns in neuroimaging data. However, most contemporary computer-assisted ASD diagnosis methods ignore the aforementioned age-/sex-related patterns. In this paper, we propose a novel sparse multiview task-centralized (Sparse-MVTC) ensemble classification method for image-based ASD diagnosis. Specifically, with the age and sex information of each subject, we formulate the classification as a multitask learning problem, where each task corresponds to learning upon a specific age/sex group. We also extract multiview features per subject to better reveal the FC changes. Then, in Sparse-MVTC learning, we select a certain central task and treat the rest as auxiliary tasks. By considering both task-task and view-view relationships between the central task and each auxiliary task, we can learn better upon the entire dataset. Finally, by selecting the central task, in turn, we are able to derive multiple classifiers for each task/group. An ensemble strategy is further adopted, such that the final diagnosis can be integrated for each subject. Our comprehensive experiments on the ABIDE database demonstrate that our proposed Sparse-MVTC ensemble learning can significantly outperform the state-of-the-art classification methods for ASD diagnosis.

Original languageEnglish
JournalIEEE Transactions on Cybernetics
DOIs
Publication statusAccepted/In press - 2018 Jun 18

Fingerprint

Neuroimaging
Brain
Classifiers
Experiments

Keywords

  • ABIDE
  • autism spectrum disorder (ASD)
  • Correlation
  • Data mining
  • diagnosis
  • Feature extraction
  • high-order functional connectivity (FC)
  • Learning systems
  • machine learning
  • Media
  • multiview multitask (MVMT) learning
  • Radiology
  • sparse multiview task-centralized (Sparse-MVTC) learning
  • Task analysis

ASJC Scopus subject areas

  • Software
  • Control and Systems Engineering
  • Information Systems
  • Human-Computer Interaction
  • Computer Science Applications
  • Electrical and Electronic Engineering

Cite this

Sparse Multiview Task-Centralized Ensemble Learning for ASD Diagnosis Based on Age- and Sex-Related Functional Connectivity Patterns. / Wang, Jun; Wang, Qian; Zhang, Han; Chen, Jiawei; Wang, Shitong; Shen, Dinggang.

In: IEEE Transactions on Cybernetics, 18.06.2018.

Research output: Contribution to journalArticle

@article{0c5eebc46eeb46c09ce2843fa22e67a8,
title = "Sparse Multiview Task-Centralized Ensemble Learning for ASD Diagnosis Based on Age- and Sex-Related Functional Connectivity Patterns",
abstract = "Autism spectrum disorder (ASD) is an age- and sex-related neurodevelopmental disorder that alters the brain's functional connectivity (FC). The changes caused by ASD are associated with different age- and sex-related patterns in neuroimaging data. However, most contemporary computer-assisted ASD diagnosis methods ignore the aforementioned age-/sex-related patterns. In this paper, we propose a novel sparse multiview task-centralized (Sparse-MVTC) ensemble classification method for image-based ASD diagnosis. Specifically, with the age and sex information of each subject, we formulate the classification as a multitask learning problem, where each task corresponds to learning upon a specific age/sex group. We also extract multiview features per subject to better reveal the FC changes. Then, in Sparse-MVTC learning, we select a certain central task and treat the rest as auxiliary tasks. By considering both task-task and view-view relationships between the central task and each auxiliary task, we can learn better upon the entire dataset. Finally, by selecting the central task, in turn, we are able to derive multiple classifiers for each task/group. An ensemble strategy is further adopted, such that the final diagnosis can be integrated for each subject. Our comprehensive experiments on the ABIDE database demonstrate that our proposed Sparse-MVTC ensemble learning can significantly outperform the state-of-the-art classification methods for ASD diagnosis.",
keywords = "ABIDE, autism spectrum disorder (ASD), Correlation, Data mining, diagnosis, Feature extraction, high-order functional connectivity (FC), Learning systems, machine learning, Media, multiview multitask (MVMT) learning, Radiology, sparse multiview task-centralized (Sparse-MVTC) learning, Task analysis",
author = "Jun Wang and Qian Wang and Han Zhang and Jiawei Chen and Shitong Wang and Dinggang Shen",
year = "2018",
month = "6",
day = "18",
doi = "10.1109/TCYB.2018.2839693",
language = "English",
journal = "IEEE Transactions on Cybernetics",
issn = "2168-2267",
publisher = "IEEE Advancing Technology for Humanity",

}

TY - JOUR

T1 - Sparse Multiview Task-Centralized Ensemble Learning for ASD Diagnosis Based on Age- and Sex-Related Functional Connectivity Patterns

AU - Wang, Jun

AU - Wang, Qian

AU - Zhang, Han

AU - Chen, Jiawei

AU - Wang, Shitong

AU - Shen, Dinggang

PY - 2018/6/18

Y1 - 2018/6/18

N2 - Autism spectrum disorder (ASD) is an age- and sex-related neurodevelopmental disorder that alters the brain's functional connectivity (FC). The changes caused by ASD are associated with different age- and sex-related patterns in neuroimaging data. However, most contemporary computer-assisted ASD diagnosis methods ignore the aforementioned age-/sex-related patterns. In this paper, we propose a novel sparse multiview task-centralized (Sparse-MVTC) ensemble classification method for image-based ASD diagnosis. Specifically, with the age and sex information of each subject, we formulate the classification as a multitask learning problem, where each task corresponds to learning upon a specific age/sex group. We also extract multiview features per subject to better reveal the FC changes. Then, in Sparse-MVTC learning, we select a certain central task and treat the rest as auxiliary tasks. By considering both task-task and view-view relationships between the central task and each auxiliary task, we can learn better upon the entire dataset. Finally, by selecting the central task, in turn, we are able to derive multiple classifiers for each task/group. An ensemble strategy is further adopted, such that the final diagnosis can be integrated for each subject. Our comprehensive experiments on the ABIDE database demonstrate that our proposed Sparse-MVTC ensemble learning can significantly outperform the state-of-the-art classification methods for ASD diagnosis.

AB - Autism spectrum disorder (ASD) is an age- and sex-related neurodevelopmental disorder that alters the brain's functional connectivity (FC). The changes caused by ASD are associated with different age- and sex-related patterns in neuroimaging data. However, most contemporary computer-assisted ASD diagnosis methods ignore the aforementioned age-/sex-related patterns. In this paper, we propose a novel sparse multiview task-centralized (Sparse-MVTC) ensemble classification method for image-based ASD diagnosis. Specifically, with the age and sex information of each subject, we formulate the classification as a multitask learning problem, where each task corresponds to learning upon a specific age/sex group. We also extract multiview features per subject to better reveal the FC changes. Then, in Sparse-MVTC learning, we select a certain central task and treat the rest as auxiliary tasks. By considering both task-task and view-view relationships between the central task and each auxiliary task, we can learn better upon the entire dataset. Finally, by selecting the central task, in turn, we are able to derive multiple classifiers for each task/group. An ensemble strategy is further adopted, such that the final diagnosis can be integrated for each subject. Our comprehensive experiments on the ABIDE database demonstrate that our proposed Sparse-MVTC ensemble learning can significantly outperform the state-of-the-art classification methods for ASD diagnosis.

KW - ABIDE

KW - autism spectrum disorder (ASD)

KW - Correlation

KW - Data mining

KW - diagnosis

KW - Feature extraction

KW - high-order functional connectivity (FC)

KW - Learning systems

KW - machine learning

KW - Media

KW - multiview multitask (MVMT) learning

KW - Radiology

KW - sparse multiview task-centralized (Sparse-MVTC) learning

KW - Task analysis

UR - http://www.scopus.com/inward/record.url?scp=85048876305&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85048876305&partnerID=8YFLogxK

U2 - 10.1109/TCYB.2018.2839693

DO - 10.1109/TCYB.2018.2839693

M3 - Article

C2 - 29994137

AN - SCOPUS:85048876305

JO - IEEE Transactions on Cybernetics

JF - IEEE Transactions on Cybernetics

SN - 2168-2267

ER -