Group ICA for identifying biomarkers in schizophrenia: 'Adaptive' networks via spatially constrained ICA show more sensitivity to group differences than spatio-temporal regression.
Classification
Functional network
GIG-ICA
ICA
Schizophrenia
Spatio-temporal regression
fMRI
Journal
NeuroImage. Clinical
ISSN: 2213-1582
Titre abrégé: Neuroimage Clin
Pays: Netherlands
ID NLM: 101597070
Informations de publication
Date de publication:
2019
2019
Historique:
received:
23
06
2018
revised:
22
01
2019
accepted:
02
03
2019
pubmed:
29
3
2019
medline:
27
12
2019
entrez:
29
3
2019
Statut:
ppublish
Résumé
Brain functional networks identified from fMRI data can provide potential biomarkers for brain disorders. Group independent component analysis (GICA) is popular for extracting brain functional networks from multiple subjects. In GICA, different strategies exist for reconstructing subject-specific networks from the group-level networks. However, it is unknown whether these strategies have different sensitivities to group differences and abilities in distinguishing patients. Among GICA, spatio-temporal regression (STR) and spatially constrained ICA approaches such as group information guided ICA (GIG-ICA) can be used to propagate components (indicating networks) to a new subject that is not included in the original subjects. In this study, based on the same a priori network maps, we reconstructed subject-specific networks using these two methods separately from resting-state fMRI data of 151 schizophrenia patients (SZs) and 163 healthy controls (HCs). We investigated group differences in the estimated functional networks and the functional network connectivity (FNC) obtained by each method. The networks were also used as features in a cross-validated support vector machine (SVM) for classifying SZs and HCs. We selected features using different strategies to provide a comprehensive comparison between the two methods. GIG-ICA generally showed greater sensitivity in statistical analysis and better classification performance (accuracy 76.45 ± 8.9%, sensitivity 0.74 ± 0.11, specificity 0.79 ± 0.11) than STR (accuracy 67.45 ± 8.13%, sensitivity 0.65 ± 0.11, specificity 0.71 ± 0.11). Importantly, results were also consistent when applied to an independent dataset including 82 HCs and 82 SZs. Our work suggests that the functional networks estimated by GIG-ICA are more sensitive to group differences, and GIG-ICA is promising for identifying image-derived biomarkers of brain disease.
Identifiants
pubmed: 30921608
pii: S2213-1582(19)30097-X
doi: 10.1016/j.nicl.2019.101747
pmc: PMC6438914
pii:
doi:
Types de publication
Comparative Study
Journal Article
Research Support, N.I.H., Extramural
Research Support, Non-U.S. Gov't
Langues
eng
Sous-ensembles de citation
IM
Pagination
101747Subventions
Organisme : NIGMS NIH HHS
ID : P20 GM103472
Pays : United States
Organisme : NCRR NIH HHS
ID : P20 RR021938
Pays : United States
Organisme : NIBIB NIH HHS
ID : R01 EB020407
Pays : United States
Organisme : NIMH NIH HHS
ID : R56 MH117107
Pays : United States
Informations de copyright
Copyright © 2019 The Authors. Published by Elsevier Inc. All rights reserved.
Références
Neural Comput. 1995 Nov;7(6):1129-59
pubmed: 7584893
Neuroimage Clin. 2017 Oct 19;17:335-346
pubmed: 29159045
IEEE Rev Biomed Eng. 2012;5:60-73
pubmed: 23231989
Neuroimage. 2013 Apr 1;69:157-97
pubmed: 23194820
Magn Reson Med. 2011 Sep;66(3):644-57
pubmed: 21394769
J Magn Reson Imaging. 2004 Mar;19(3):365-8
pubmed: 14994306
Neuroimage. 2012 Feb 15;59(4):4141-59
pubmed: 22019879
Conf Proc IEEE Eng Med Biol Soc. 2014;2014:1517-20
pubmed: 25570258
Hum Brain Mapp. 2008 Nov;29(11):1265-75
pubmed: 17894392
Acta Psychiatr Scand. 2018 Nov;138(5):472-482
pubmed: 30084192
IEEE Trans Med Imaging. 2002 May;21(5):470-84
pubmed: 12071618
AJNR Am J Neuroradiol. 2001 Aug;22(7):1326-33
pubmed: 11498421
Neuroimage Clin. 2014 Jul 24;5:298-308
pubmed: 25161896
Biol Psychiatry. 2012 May 15;71(10):881-9
pubmed: 22401986
Med Image Comput Comput Assist Interv. 2018 Sep;11072:293-301
pubmed: 31106304
Hum Brain Mapp. 2017 Dec;38(12):5845-5858
pubmed: 28876500
Hum Brain Mapp. 2011 Dec;32(12):2075-95
pubmed: 21162045
Front Hum Neurosci. 2012 Jun 04;6:145
pubmed: 22675292
Neuroimage. 2015 Nov 15;122:272-80
pubmed: 26216278
Front Syst Neurosci. 2011 Feb 04;5:2
pubmed: 21442040
Hum Brain Mapp. 1998;6(3):160-88
pubmed: 9673671
Front Hum Neurosci. 2014 Nov 07;8:897
pubmed: 25426048
Hum Brain Mapp. 2016 Mar;37(3):1005-25
pubmed: 26859308
Neuroimage. 2014 Nov 15;102 Pt 2:620-36
pubmed: 25150630
Schizophr Bull. 2017 Jul 1;43(4):914-924
pubmed: 27872268
Neuroimage. 2006 Nov 1;33(2):471-81
pubmed: 16952468
Hum Brain Mapp. 2010 Jul;31(7):1076-88
pubmed: 20017117
Biol Psychiatry. 2010 Feb 1;67(3):255-62
pubmed: 19897178
Comput Biomed Res. 1996 Jun;29(3):162-73
pubmed: 8812068
Prog Neuropsychopharmacol Biol Psychiatry. 2019 Jan 10;88:66-73
pubmed: 29935206
Neuroimage. 2016 Jan 1;124(Pt B):1074-1079
pubmed: 26364863
Neuroimage. 2013 Oct 15;80:144-68
pubmed: 23702415
NPJ Schizophr. 2015 May 06;1:15016
pubmed: 27336032
Neuroimage. 2008 Mar 1;40(1):86-109
pubmed: 18165105
Front Syst Neurosci. 2014 Jun 26;8:106
pubmed: 25018704
Neuroimage. 2011 Sep 15;58(2):526-36
pubmed: 21723948
Elife. 2018 Feb 16;7:
pubmed: 29451491
Neuroimage Clin. 2018 May 09;19:476-486
pubmed: 29984156
Schizophr Res. 2016 Jan;170(1):55-65
pubmed: 26654933
Neuroimage. 2005 Mar;25(1):294-311
pubmed: 15734364
Neuroimage. 2002 Jul;16(3 Pt 1):551-63
pubmed: 12169242
IEEE Trans Biomed Eng. 2011 Dec;58(12):3406-17
pubmed: 21900068
Hum Brain Mapp. 2001 Nov;14(3):140-51
pubmed: 11559959
Front Neurosci. 2017 May 19;11:267
pubmed: 28579940
Neuroimage. 2014 Jul 15;95:232-47
pubmed: 24657355
Neuroimage. 2018 Feb 15;167:62-72
pubmed: 29155080
Neuroimage. 2017 Jan 15;145(Pt B):137-165
pubmed: 27012503
Front Neurosci. 2018 Aug 06;12:525
pubmed: 30127711