Parallel group ICA+ICA: Joint estimation of linked functional network variability and structural covariation with application to schizophrenia

Shile Qi, Jing Sui, Jiayu Chen, Jingyu Liu, Rongtao Jiang, Rogers Silva, Armin Iraji, Eswar Damaraju, Mustafa Salman, Dongdong Lin, Zening Fu, Dongmei Zhi, Jessica A. Turner, Juan Bustillo, Judith M. Ford, Daniel H. Mathalon, James Voyvodic, Sarah McEwen, Adrian Preda, Aysenil BelgerSteven G. Potkin, Bryon A. Mueller, Tulay Adali, Vince Daniel Calhoun

Research output: Contribution to journalArticle

Abstract

There is growing evidence that rather than using a single brain imaging modality to study its association with physiological or symptomatic features, the field is paying more attention to fusion of multimodal information. However, most current multimodal fusion approaches that incorporate functional magnetic resonance imaging (fMRI) are restricted to second-level 3D features, rather than the original 4D fMRI data. This trade-off is that the valuable temporal information is not utilized during the fusion step. Here we are motivated to propose a novel approach called “parallel group ICA+ICA” that incorporates temporal fMRI information from group independent component analysis (GICA) into a parallel independent component analysis (ICA) framework, aiming to enable direct fusion of first-level fMRI features with other modalities (e.g., structural MRI), which thus can detect linked functional network variability and structural covariations. Simulation results show that the proposed method yields accurate intermodality linkage detection regardless of whether it is strong or weak. When applied to real data, we identified one pair of significantly associated fMRI-sMRI components that show group difference between schizophrenia and controls in both modalities, and this linkage can be replicated in an independent cohort. Finally, multiple cognitive domain scores can be predicted by the features identified in the linked component pair by our proposed method. We also show these multimodal brain features can predict multiple cognitive scores in an independent cohort. Overall, results demonstrate the ability of parallel GICA+ICA to estimate joint information from 4D and 3D data without discarding much of the available information up front, and the potential for using this approach to identify imaging biomarkers to study brain disorders.

Original languageEnglish (US)
JournalHuman Brain Mapping
DOIs
StatePublished - Jan 1 2019
Externally publishedYes

Fingerprint

Schizophrenia
Joints
Magnetic Resonance Imaging
Aptitude
Brain Diseases
Neuroimaging
Biomarkers
Brain

Keywords

  • group independent component analysis
  • multimodal fusion
  • parallel independent component analysis
  • schizophrenia
  • subjects' variability
  • temporal information

ASJC Scopus subject areas

  • Anatomy
  • Radiological and Ultrasound Technology
  • Radiology Nuclear Medicine and imaging
  • Neurology
  • Clinical Neurology

Cite this

Parallel group ICA+ICA : Joint estimation of linked functional network variability and structural covariation with application to schizophrenia. / Qi, Shile; Sui, Jing; Chen, Jiayu; Liu, Jingyu; Jiang, Rongtao; Silva, Rogers; Iraji, Armin; Damaraju, Eswar; Salman, Mustafa; Lin, Dongdong; Fu, Zening; Zhi, Dongmei; Turner, Jessica A.; Bustillo, Juan; Ford, Judith M.; Mathalon, Daniel H.; Voyvodic, James; McEwen, Sarah; Preda, Adrian; Belger, Aysenil; Potkin, Steven G.; Mueller, Bryon A.; Adali, Tulay; Calhoun, Vince Daniel.

In: Human Brain Mapping, 01.01.2019.

Research output: Contribution to journalArticle

Qi, S, Sui, J, Chen, J, Liu, J, Jiang, R, Silva, R, Iraji, A, Damaraju, E, Salman, M, Lin, D, Fu, Z, Zhi, D, Turner, JA, Bustillo, J, Ford, JM, Mathalon, DH, Voyvodic, J, McEwen, S, Preda, A, Belger, A, Potkin, SG, Mueller, BA, Adali, T & Calhoun, VD 2019, 'Parallel group ICA+ICA: Joint estimation of linked functional network variability and structural covariation with application to schizophrenia', Human Brain Mapping. https://doi.org/10.1002/hbm.24632
Qi, Shile ; Sui, Jing ; Chen, Jiayu ; Liu, Jingyu ; Jiang, Rongtao ; Silva, Rogers ; Iraji, Armin ; Damaraju, Eswar ; Salman, Mustafa ; Lin, Dongdong ; Fu, Zening ; Zhi, Dongmei ; Turner, Jessica A. ; Bustillo, Juan ; Ford, Judith M. ; Mathalon, Daniel H. ; Voyvodic, James ; McEwen, Sarah ; Preda, Adrian ; Belger, Aysenil ; Potkin, Steven G. ; Mueller, Bryon A. ; Adali, Tulay ; Calhoun, Vince Daniel. / Parallel group ICA+ICA : Joint estimation of linked functional network variability and structural covariation with application to schizophrenia. In: Human Brain Mapping. 2019.
@article{ddb8767c5f9348bc99f6f5f2d1d4cf5c,
title = "Parallel group ICA+ICA: Joint estimation of linked functional network variability and structural covariation with application to schizophrenia",
abstract = "There is growing evidence that rather than using a single brain imaging modality to study its association with physiological or symptomatic features, the field is paying more attention to fusion of multimodal information. However, most current multimodal fusion approaches that incorporate functional magnetic resonance imaging (fMRI) are restricted to second-level 3D features, rather than the original 4D fMRI data. This trade-off is that the valuable temporal information is not utilized during the fusion step. Here we are motivated to propose a novel approach called “parallel group ICA+ICA” that incorporates temporal fMRI information from group independent component analysis (GICA) into a parallel independent component analysis (ICA) framework, aiming to enable direct fusion of first-level fMRI features with other modalities (e.g., structural MRI), which thus can detect linked functional network variability and structural covariations. Simulation results show that the proposed method yields accurate intermodality linkage detection regardless of whether it is strong or weak. When applied to real data, we identified one pair of significantly associated fMRI-sMRI components that show group difference between schizophrenia and controls in both modalities, and this linkage can be replicated in an independent cohort. Finally, multiple cognitive domain scores can be predicted by the features identified in the linked component pair by our proposed method. We also show these multimodal brain features can predict multiple cognitive scores in an independent cohort. Overall, results demonstrate the ability of parallel GICA+ICA to estimate joint information from 4D and 3D data without discarding much of the available information up front, and the potential for using this approach to identify imaging biomarkers to study brain disorders.",
keywords = "group independent component analysis, multimodal fusion, parallel independent component analysis, schizophrenia, subjects' variability, temporal information",
author = "Shile Qi and Jing Sui and Jiayu Chen and Jingyu Liu and Rongtao Jiang and Rogers Silva and Armin Iraji and Eswar Damaraju and Mustafa Salman and Dongdong Lin and Zening Fu and Dongmei Zhi and Turner, {Jessica A.} and Juan Bustillo and Ford, {Judith M.} and Mathalon, {Daniel H.} and James Voyvodic and Sarah McEwen and Adrian Preda and Aysenil Belger and Potkin, {Steven G.} and Mueller, {Bryon A.} and Tulay Adali and Calhoun, {Vince Daniel}",
year = "2019",
month = "1",
day = "1",
doi = "10.1002/hbm.24632",
language = "English (US)",
journal = "Human Brain Mapping",
issn = "1065-9471",
publisher = "Wiley-Liss Inc.",

}

TY - JOUR

T1 - Parallel group ICA+ICA

T2 - Joint estimation of linked functional network variability and structural covariation with application to schizophrenia

AU - Qi, Shile

AU - Sui, Jing

AU - Chen, Jiayu

AU - Liu, Jingyu

AU - Jiang, Rongtao

AU - Silva, Rogers

AU - Iraji, Armin

AU - Damaraju, Eswar

AU - Salman, Mustafa

AU - Lin, Dongdong

AU - Fu, Zening

AU - Zhi, Dongmei

AU - Turner, Jessica A.

AU - Bustillo, Juan

AU - Ford, Judith M.

AU - Mathalon, Daniel H.

AU - Voyvodic, James

AU - McEwen, Sarah

AU - Preda, Adrian

AU - Belger, Aysenil

AU - Potkin, Steven G.

AU - Mueller, Bryon A.

AU - Adali, Tulay

AU - Calhoun, Vince Daniel

PY - 2019/1/1

Y1 - 2019/1/1

N2 - There is growing evidence that rather than using a single brain imaging modality to study its association with physiological or symptomatic features, the field is paying more attention to fusion of multimodal information. However, most current multimodal fusion approaches that incorporate functional magnetic resonance imaging (fMRI) are restricted to second-level 3D features, rather than the original 4D fMRI data. This trade-off is that the valuable temporal information is not utilized during the fusion step. Here we are motivated to propose a novel approach called “parallel group ICA+ICA” that incorporates temporal fMRI information from group independent component analysis (GICA) into a parallel independent component analysis (ICA) framework, aiming to enable direct fusion of first-level fMRI features with other modalities (e.g., structural MRI), which thus can detect linked functional network variability and structural covariations. Simulation results show that the proposed method yields accurate intermodality linkage detection regardless of whether it is strong or weak. When applied to real data, we identified one pair of significantly associated fMRI-sMRI components that show group difference between schizophrenia and controls in both modalities, and this linkage can be replicated in an independent cohort. Finally, multiple cognitive domain scores can be predicted by the features identified in the linked component pair by our proposed method. We also show these multimodal brain features can predict multiple cognitive scores in an independent cohort. Overall, results demonstrate the ability of parallel GICA+ICA to estimate joint information from 4D and 3D data without discarding much of the available information up front, and the potential for using this approach to identify imaging biomarkers to study brain disorders.

AB - There is growing evidence that rather than using a single brain imaging modality to study its association with physiological or symptomatic features, the field is paying more attention to fusion of multimodal information. However, most current multimodal fusion approaches that incorporate functional magnetic resonance imaging (fMRI) are restricted to second-level 3D features, rather than the original 4D fMRI data. This trade-off is that the valuable temporal information is not utilized during the fusion step. Here we are motivated to propose a novel approach called “parallel group ICA+ICA” that incorporates temporal fMRI information from group independent component analysis (GICA) into a parallel independent component analysis (ICA) framework, aiming to enable direct fusion of first-level fMRI features with other modalities (e.g., structural MRI), which thus can detect linked functional network variability and structural covariations. Simulation results show that the proposed method yields accurate intermodality linkage detection regardless of whether it is strong or weak. When applied to real data, we identified one pair of significantly associated fMRI-sMRI components that show group difference between schizophrenia and controls in both modalities, and this linkage can be replicated in an independent cohort. Finally, multiple cognitive domain scores can be predicted by the features identified in the linked component pair by our proposed method. We also show these multimodal brain features can predict multiple cognitive scores in an independent cohort. Overall, results demonstrate the ability of parallel GICA+ICA to estimate joint information from 4D and 3D data without discarding much of the available information up front, and the potential for using this approach to identify imaging biomarkers to study brain disorders.

KW - group independent component analysis

KW - multimodal fusion

KW - parallel independent component analysis

KW - schizophrenia

KW - subjects' variability

KW - temporal information

UR - http://www.scopus.com/inward/record.url?scp=85066038485&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85066038485&partnerID=8YFLogxK

U2 - 10.1002/hbm.24632

DO - 10.1002/hbm.24632

M3 - Article

C2 - 31099151

AN - SCOPUS:85066038485

JO - Human Brain Mapping

JF - Human Brain Mapping

SN - 1065-9471

ER -