Table 4.
Study | Network type | Subjects | Node definition | Edge definition | Analysis | Global results* | Local Results* | Hubs* |
---|---|---|---|---|---|---|---|---|
Bassett et al., 2008 | Structural correlation (1.5 T) | 203 SCZ | 108 regions Pick atlas | Binary, correlation between region volumes | Graph theory | =SW, K distribution | ↓hierarchy, ↑mean C distance in multimodal network | ↓PFC hubs |
259 HC | ≠ CC, K and B in PreM, PFC, OFC, ITG, MTG, CG, Ins | ↑ITG, Ins, CG hubs | ||||||
Zhang et al., 2012b | Structural correlation (1.5 T) | 101 SCZ | 78 regions AAL atlas | Binary, correlation between regional cortical thickness | Graph theory | ↑PL, CC | ↓B mainly in the association cortices | ↑# of hubs |
101 HC | ↑B in the primary and paralimbic cortices | ↓association cortex hubs | ||||||
↑primary, paralimbic hubs | ||||||||
Collin et al., 2013 | Structural correlation (1.5T) | 146 SCZ | (1) 12 lobes | Binary, correlation between region volumes | C compare (lobes) | n/a | ↓C frontal <-> subcortical | n/a |
122 HC | (2) 82 regions Freesurfer | NBS (regions) | ↑C temporal <-> subcortical; L frontal <-> R frontal; frontal <-> limbic impaired networks:1) | |||||
PFC, temporal, occipital, parietal, limbic 2) SMG, Post-CG, CG, subcortical (NBS) | ||||||||
Shi et al., 2012 | Structural correlation (3T) | 26 high risk neonates$ | 90 regions adapted AAL atlas | Binary, correlation between region volumes | Graph theory | ↓Eglob, inter-module C | n/a | =frontal, temporal hubs |
26 HC neonates | ↑Eloc, PL, CC, C distance | ↑occipital hubs | ||||||
↓parietal, subcortical hubs | ||||||||
Shi et al., 2012 | DTI white matter (3T, 6 dir) | 26 high risk neonates | 90 regions adapted AAL atlas | Binary, nodes linked if sufficient # of streamlines | Graph theory | =Eglob, Eloc, C distance | ↓number of fibers in 4 right subcortical <-> cortical and 2 occipital <-> limbic Cs | =frontal, temporal, insula, limbic hubs |
26 HC neonates | C compare | ↑occipital hubs | ||||||
↓parietal, subcortical hubs | ||||||||
van den Heuvel et al., 2010 | DTI white matter (1.5T, 32dir) | 40 SCZ | 108 regions AAL atlas | Weighted by average MTR -nodes linked if sufficient # of streamlines | Graph theory | =SW, C strength, PL, CC | ↑PL of frontal and temporal regions | ↓B in frontal hubs |
40 HC | ↓CC in HPC and paracentral lobule | |||||||
Skudlarski et al., 2010 | DTI White matter (3T, 12dir) | 27 SCZ | 6000 K means clustering | Weighted by # of streamlines | C compare | Decoupling between anatomical and functional C | ↓C in in default mode and task positive networks | n/a |
27 HC | ||||||||
Zalesky et al., 2011 | DTI white matter (1.5T, 64dir) | 74 SCZ | 82 regions AAL atlas | Binary, nodes linked if sufficient # of streamlines | Graph theory | =SW | ↓C in AG <-> R MTG; R SFG <-> L SFG; L IFG <-> L Ins | n/a |
32 HC | C compare | ↓K, Eglob | ↓C in network of medial frontal regions connected to parietal and occipital lobes (NBS) | |||||
NBS | ||||||||
Wang et al., 2012 | DTI white matter (1.5T, 13dir) | 79 SCZ | 90 regions AAL atlas | Weighted by # of streamlines | Graph theory | ↓Eglob | ↓regional Eglob in frontal associative cortices, paralimbic/limbic and subcortical regions | =association cortex, paralimbic/ limbic hubs |
96 HC | =Eloc, SW | |||||||
van den Heuvel et al., 2013 | DTI white matter (3T, 32dir) | 48 SCZ | 82 regions | Weighted by # of streamlines | Graph theory | ↓RC organization, C strength, Eglob, CC | ↓RC organization most pronounced in cortical networks | =RC members: PCUN, SFG, SPG, Ins |
45 HC | Freesurfer | ↑M | ↓density of RC Cs | |||||
=SW | ||||||||
Collin et al., 2014 | DTI white matter (1.5T, 32dir) | 40 SCZ | 68 cortical regions | Weighted by # of streamlines | Graph theory | ↓C strength, Eglob (HC>SCZ) | ↓C strength: SFG, ITG ↓Eglob: ACG, SFG, and PreCG | ≠C and CC over-represented in RC members |
54 SCZ siblings | Freesurfer | NBS | ↓CC (HC>siblings>SCZ) | ↓CC: SFG, ACG, OFG, PreCG and Ins | ↓C over- represented in RC members (NBS) | |||
51 HC | ↓RC and local C strength (HC>siblings>SCZ) | |||||||
Ottet et al., 2013 | DTI white matter (3T, 30dir) | 46 22q11.2DS | 82 regions | Binary, nodes linked if sufficient # of streamlines | graph theory | ↑PL | ↓centrality in 16 out of 65 nonhubs (25%) | ↓centrality in 10 out of 17 hubs (58%): HPC, SPG and PreCG, MFG, SFG, PCUN, THAL |
48 HC | Freesurfer | ↓Eglob, K | ||||||
=CC | ||||||||
Zhang et al., 2014 | DTI white matter (1.5T, 15dir) | 30 FE-SCZ | 90 regions AAL atlas | Weighted by # of streamlines | graph theory | =SW | ↓Eglob and K in sensorimotor, basal ganglia, and limbic-visual systems | =hubs: PCUN, FG, MTG, ITG, SPG, post-CG |
34 HC | NBS | ↓C strength, Eglob, K | ↓C in subnetwork of frontal, parietal, occipital and subcortical regions (NBS) | ↓hubs: MFG, post-CG and PreCG | ||||
↑PL |
Results reported in patient groups relative to controls unless otherwise indicated;
High risk neonates have mothers with schizophrenia.
SCZ, schizophrenia; SA, schizoaffective disorder; SF, schizophreniform disorder; FE-SCZ, first episode, medication naïve schizophrenia; HC, healthy control; 22q11.2DS - 22q11.2 deletion syndrome; NBS, network based statistics; MTR, magnetization transfer ratio; AAL, Automated Anatomical Labeling atlas (Tzourio-Mazoyer et al., 2002); LPBA40, LONI Probabilistic Brain Atlas (Shattuck et al., 2008).
Graph theory metrics: SW, small world organization; CC, clustering coefficient; PL, mean path length; K, degree; B, betweenness centrality; Eglob, global efficiency; Eloc, local efficiency; C, connectivity; M, modularity; RC, rich club.
Symbols: <->, between; #, number; ↑, increased; ↓, decreased; =, equivalent; ≠, different.
Brain regions: AG, angular gyrus; CG, cingulate gyrus; HPC, hippocampus; IFG, inferior frontal gyrus; Ins, Insular cortex; ITG, inferior temporal gyrus; MFG, middle frontal gyrus; MTG, middle temporal gyrus; OFC, orbital frontal cortex; PCUN, precuneus; PFC, prefrontal cortex; PreCG, precentral gyrus; PreM, premotor cortex; SFG, superior frontal gyrus; SPG, superior parietal gyrus; THAL, thalamus.