Graph Convolutional Networks Reveal Network-Level Functional Dysconnectivity in Schizophrenia.
connectome
graph analysis
machine learning
magnetic resonance imaging
neuroimaging
psychosis
Journal
Schizophrenia bulletin
ISSN: 1745-1701
Titre abrégé: Schizophr Bull
Pays: United States
ID NLM: 0236760
Informations de publication
Date de publication:
21 06 2022
21 06 2022
Historique:
pubmed:
16
5
2022
medline:
24
6
2022
entrez:
15
5
2022
Statut:
ppublish
Résumé
Schizophrenia is increasingly understood as a disorder of brain dysconnectivity. Recently, graph-based approaches such as graph convolutional network (GCN) have been leveraged to explore complex pairwise similarities in imaging features among brain regions, which can reveal abstract and complex relationships within brain networks. We used GCN to investigate topological abnormalities of functional brain networks in schizophrenia. Resting-state functional magnetic resonance imaging data were acquired from 505 individuals with schizophrenia and 907 controls across 6 sites. Whole-brain functional connectivity matrix was extracted for each individual. We examined the performance of GCN relative to support vector machine (SVM), extracted the most salient regions contributing to both classification models, investigated the topological profiles of identified salient regions, and explored correlation between nodal topological properties of each salient region and severity of symptom. GCN enabled nominally higher classification accuracy (85.8%) compared with SVM (80.9%). Based on the saliency map, the most discriminative brain regions were located in a distributed network including striatal areas (ie, putamen, pallidum, and caudate) and the amygdala. Significant differences in the nodal efficiency of bilateral putamen and pallidum between patients and controls and its correlations with negative symptoms were detected in post hoc analysis. The present study demonstrates that GCN allows classification of schizophrenia at the individual level with high accuracy, indicating a promising direction for detection of individual patients with schizophrenia. Functional topological deficits of striatal areas may represent a focal neural deficit of negative symptomatology in schizophrenia.
Sections du résumé
BACKGROUND AND HYPOTHESIS
Schizophrenia is increasingly understood as a disorder of brain dysconnectivity. Recently, graph-based approaches such as graph convolutional network (GCN) have been leveraged to explore complex pairwise similarities in imaging features among brain regions, which can reveal abstract and complex relationships within brain networks.
STUDY DESIGN
We used GCN to investigate topological abnormalities of functional brain networks in schizophrenia. Resting-state functional magnetic resonance imaging data were acquired from 505 individuals with schizophrenia and 907 controls across 6 sites. Whole-brain functional connectivity matrix was extracted for each individual. We examined the performance of GCN relative to support vector machine (SVM), extracted the most salient regions contributing to both classification models, investigated the topological profiles of identified salient regions, and explored correlation between nodal topological properties of each salient region and severity of symptom.
STUDY RESULTS
GCN enabled nominally higher classification accuracy (85.8%) compared with SVM (80.9%). Based on the saliency map, the most discriminative brain regions were located in a distributed network including striatal areas (ie, putamen, pallidum, and caudate) and the amygdala. Significant differences in the nodal efficiency of bilateral putamen and pallidum between patients and controls and its correlations with negative symptoms were detected in post hoc analysis.
CONCLUSIONS
The present study demonstrates that GCN allows classification of schizophrenia at the individual level with high accuracy, indicating a promising direction for detection of individual patients with schizophrenia. Functional topological deficits of striatal areas may represent a focal neural deficit of negative symptomatology in schizophrenia.
Identifiants
pubmed: 35569019
pii: 6586020
doi: 10.1093/schbul/sbac047
pmc: PMC9212102
doi:
Types de publication
Journal Article
Research Support, Non-U.S. Gov't
Langues
eng
Sous-ensembles de citation
IM
Pagination
881-892Subventions
Organisme : Wellcome Trust
Pays : United Kingdom
Organisme : Medical Research Council
ID : MC_G0802534
Pays : United Kingdom
Organisme : Wellcome Trust
ID : 220402/Z/20/Z
Pays : United Kingdom
Organisme : Wellcome Trust
ID : WT213038/Z/18/Z
Pays : United Kingdom
Informations de copyright
© The Author(s) 2022. Published by Oxford University Press on behalf of the Maryland Psychiatric Research Center.
Références
Neuropsychol Rev. 2014 Mar;24(1):32-48
pubmed: 24500505
JAMA Psychiatry. 2015 Apr;72(4):305-15
pubmed: 25651064
Am J Psychiatry. 2003 Jan;160(1):13-23
pubmed: 12505794
Am J Psychiatry. 1991 Nov;148(11):1474-86
pubmed: 1681750
Neuroimage Clin. 2021;29:102531
pubmed: 33340977
Biol Psychiatry Cogn Neurosci Neuroimaging. 2021 Dec;6(12):1125-1134
pubmed: 32800754
Schizophr Bull. 1976;2(1):19-76
pubmed: 779020
Neuroimage. 2009 Mar;45(1 Suppl):S199-209
pubmed: 19070668
Hum Brain Mapp. 2020 Apr 1;41(5):1119-1135
pubmed: 31737978
J Neurosci. 2010 Jul 14;30(28):9477-87
pubmed: 20631176
Annu Rev Clin Psychol. 2011;7:113-40
pubmed: 21128784
Front Syst Neurosci. 2011 Feb 08;5:7
pubmed: 21369355
EBioMedicine. 2018 Apr;30:74-85
pubmed: 29622496
Brain Imaging Behav. 2019 Oct;13(5):1220-1235
pubmed: 30094555
Trends Neurosci. 2019 Mar;42(3):205-220
pubmed: 30621912
Am J Manag Care. 2014 Jun;20(8 Suppl):S160-5
pubmed: 25180705
BMC Neurosci. 2022 Jan 17;23(1):5
pubmed: 35038994
J Psychiatry Neurosci. 2018 Nov 01;43(6):427
pubmed: 30375837
Schizophr Res. 2016 Aug;175(1-3):20-26
pubmed: 27102424
Biol Psychiatry Cogn Neurosci Neuroimaging. 2020 Aug;5(8):799-807
pubmed: 31447329
Hum Brain Mapp. 2018 Nov;39(11):4213-4227
pubmed: 29962049
Science. 2010 Sep 10;329(5997):1358-61
pubmed: 20829489
Hum Brain Mapp. 2020 Dec;41(17):4997-5014
pubmed: 32813309
Psychol Med. 2019 Nov;49(15):2533-2542
pubmed: 30460891
Eur Neuropsychopharmacol. 2015 May;25(5):733-48
pubmed: 24726580
Nat Rev Neurosci. 2009 Mar;10(3):186-98
pubmed: 19190637
Front Psychiatry. 2017 Nov 20;8:245
pubmed: 29209237
Schizophr Bull. 2020 Jan 4;46(1):17-26
pubmed: 30809667
Am J Psychiatry. 2013 Jan;170(1):59-70
pubmed: 23111466
Front Hum Neurosci. 2015 Jun 30;9:386
pubmed: 26175682
Med Image Anal. 2018 Aug;48:117-130
pubmed: 29890408
PLoS Comput Biol. 2005 Sep;1(4):e42
pubmed: 16201007
Schizophr Res. 2015 Dec;169(1-3):178-185
pubmed: 26527246
Adv Neural Inf Process Syst. 2019 Dec;32:9240-9251
pubmed: 32265580
Curr Opin Psychiatry. 2015 May;28(3):229-35
pubmed: 25768086
J Neurosci. 2010 Nov 24;30(47):15915-26
pubmed: 21106830
Brain Imaging Behav. 2020 Dec;14(6):2232-2241
pubmed: 31376115
Arch Gen Psychiatry. 1988 Nov;45(11):1055
pubmed: 3178414
Front Syst Neurosci. 2010 May 14;4:13
pubmed: 20577591
Psychol Med. 2021 Jan;51(2):340-350
pubmed: 31858920
Dialogues Clin Neurosci. 2018 Jun;20(2):111-121
pubmed: 30250388
Neuroimage. 2018 Apr 1;169:431-442
pubmed: 29278772
Proc Natl Acad Sci U S A. 2019 May 7;116(19):9604-9609
pubmed: 31004051
J Abnorm Psychol. 2015 Aug;124(3):764-9
pubmed: 26098046
Neuropsychiatr Dis Treat. 2019 Jun 19;15:1605-1627
pubmed: 31354276
Schizophr Res. 2002 Dec 1;58(2-3):123-34
pubmed: 12409152
Annu Rev Psychol. 2009;60:693-716
pubmed: 18771388
Mol Psychiatry. 2016 Oct;21(10):1460-6
pubmed: 26782053
Biol Psychiatry. 2015 Dec 1;78(11):794-804
pubmed: 25847180
Arch Gen Psychiatry. 1995 Dec;52(12):1061-70
pubmed: 7492258
Transl Psychiatry. 2020 Apr 20;10(1):107
pubmed: 32313006
Brain. 1995 Feb;118 ( Pt 1):15-24
pubmed: 7895001
Neuropsychopharmacology. 2019 Apr;44(5):869-875
pubmed: 30127342
Trends Cogn Sci. 2013 Dec;17(12):641-7
pubmed: 24238779
Neuroimage. 2017 Jan 15;145(Pt B):137-165
pubmed: 27012503
IEEE Trans Neural Syst Rehabil Eng. 2021;29:1784-1794
pubmed: 34406943
Neurosci Biobehav Rev. 2017 Mar;74(Pt A):58-75
pubmed: 28087243
Psychol Med. 2020 Aug;50(11):1852-1861
pubmed: 31391132
Psychiatry Res. 2001 Dec 31;105(3):265-71
pubmed: 11814545
Methods Mol Biol. 2017;1613:479-504
pubmed: 28849573
J Behav Addict. 2020 Apr 01;9(1):44-57
pubmed: 32359230