Pseudo-bootstrap network analysis—an application in functional connectivity fingerprinting

Hu Cheng, Ao Li, Andrea A. Koenigsberger, Chunfeng Huang, Yang Wang, Jinhua Sheng, Sharlene D. Newman

Research output: Contribution to journalArticle

Abstract

Brain parcellation divides the brain’s spatial domain into small regions, which are represented by nodes within the network analysis framework. While template-based parcellations are widely used, the parcels on the template do not necessarily match individual’s functional nodes. A new method is developed to overcome the inconsistent network analysis results by by-passing the difficulties of parcellating the brain into functionally meaningful areas. First, roughly equal-sized parcellations are obtained. Second, these random parcellations are applied to individual subjects multiple times and a pseudo-bootstrap (PBS) of the network is obtained for statistical inferences. It was found that the variation of mean global network metrics from PBS sampling is smaller compared with inter-subject variation or within-subject variation between two diffusion MRI scans. Using the mean global network metrics from PBS sampling, the intra-class correlation is always higher than the average obtained from using a single random parcellation. As one application, the PBS method was tested on the Human Connectome Project resting state dataset to identify individuals across scan sessions based on the mean functional connectivity (FC)—a trivial network property that has little information about the connectivity between nodes. An accuracy rate of ∼90% was achieved by simply finding the maximum correlation of mean FC of PBS samples between two scan sessions.

Original languageEnglish (US)
Article number351
JournalFrontiers in Human Neuroscience
Volume11
DOIs
StatePublished - Jul 13 2017
Externally publishedYes

Fingerprint

Brain
Connectome
Diffusion Magnetic Resonance Imaging
Magnetic Resonance Imaging
Datasets

Keywords

  • Connectomes
  • Functional connectivity fingerprint
  • Intra-class correlation coefficient
  • Network analysis
  • Pseudo-bootstrap
  • Random parcellation

ASJC Scopus subject areas

  • Neuropsychology and Physiological Psychology
  • Neurology
  • Psychiatry and Mental health
  • Biological Psychiatry
  • Behavioral Neuroscience

Cite this

Pseudo-bootstrap network analysis—an application in functional connectivity fingerprinting. / Cheng, Hu; Li, Ao; Koenigsberger, Andrea A.; Huang, Chunfeng; Wang, Yang; Sheng, Jinhua; Newman, Sharlene D.

In: Frontiers in Human Neuroscience, Vol. 11, 351, 13.07.2017.

Research output: Contribution to journalArticle

Cheng, Hu ; Li, Ao ; Koenigsberger, Andrea A. ; Huang, Chunfeng ; Wang, Yang ; Sheng, Jinhua ; Newman, Sharlene D. / Pseudo-bootstrap network analysis—an application in functional connectivity fingerprinting. In: Frontiers in Human Neuroscience. 2017 ; Vol. 11.
@article{9131bd17329344c89b335b4c4ccce37e,
title = "Pseudo-bootstrap network analysis—an application in functional connectivity fingerprinting",
abstract = "Brain parcellation divides the brain’s spatial domain into small regions, which are represented by nodes within the network analysis framework. While template-based parcellations are widely used, the parcels on the template do not necessarily match individual’s functional nodes. A new method is developed to overcome the inconsistent network analysis results by by-passing the difficulties of parcellating the brain into functionally meaningful areas. First, roughly equal-sized parcellations are obtained. Second, these random parcellations are applied to individual subjects multiple times and a pseudo-bootstrap (PBS) of the network is obtained for statistical inferences. It was found that the variation of mean global network metrics from PBS sampling is smaller compared with inter-subject variation or within-subject variation between two diffusion MRI scans. Using the mean global network metrics from PBS sampling, the intra-class correlation is always higher than the average obtained from using a single random parcellation. As one application, the PBS method was tested on the Human Connectome Project resting state dataset to identify individuals across scan sessions based on the mean functional connectivity (FC)—a trivial network property that has little information about the connectivity between nodes. An accuracy rate of ∼90{\%} was achieved by simply finding the maximum correlation of mean FC of PBS samples between two scan sessions.",
keywords = "Connectomes, Functional connectivity fingerprint, Intra-class correlation coefficient, Network analysis, Pseudo-bootstrap, Random parcellation",
author = "Hu Cheng and Ao Li and Koenigsberger, {Andrea A.} and Chunfeng Huang and Yang Wang and Jinhua Sheng and Newman, {Sharlene D.}",
year = "2017",
month = "7",
day = "13",
doi = "10.3389/fnhum.2017.00351",
language = "English (US)",
volume = "11",
journal = "Frontiers in Human Neuroscience",
issn = "1662-5161",
publisher = "Frontiers Research Foundation",

}

TY - JOUR

T1 - Pseudo-bootstrap network analysis—an application in functional connectivity fingerprinting

AU - Cheng, Hu

AU - Li, Ao

AU - Koenigsberger, Andrea A.

AU - Huang, Chunfeng

AU - Wang, Yang

AU - Sheng, Jinhua

AU - Newman, Sharlene D.

PY - 2017/7/13

Y1 - 2017/7/13

N2 - Brain parcellation divides the brain’s spatial domain into small regions, which are represented by nodes within the network analysis framework. While template-based parcellations are widely used, the parcels on the template do not necessarily match individual’s functional nodes. A new method is developed to overcome the inconsistent network analysis results by by-passing the difficulties of parcellating the brain into functionally meaningful areas. First, roughly equal-sized parcellations are obtained. Second, these random parcellations are applied to individual subjects multiple times and a pseudo-bootstrap (PBS) of the network is obtained for statistical inferences. It was found that the variation of mean global network metrics from PBS sampling is smaller compared with inter-subject variation or within-subject variation between two diffusion MRI scans. Using the mean global network metrics from PBS sampling, the intra-class correlation is always higher than the average obtained from using a single random parcellation. As one application, the PBS method was tested on the Human Connectome Project resting state dataset to identify individuals across scan sessions based on the mean functional connectivity (FC)—a trivial network property that has little information about the connectivity between nodes. An accuracy rate of ∼90% was achieved by simply finding the maximum correlation of mean FC of PBS samples between two scan sessions.

AB - Brain parcellation divides the brain’s spatial domain into small regions, which are represented by nodes within the network analysis framework. While template-based parcellations are widely used, the parcels on the template do not necessarily match individual’s functional nodes. A new method is developed to overcome the inconsistent network analysis results by by-passing the difficulties of parcellating the brain into functionally meaningful areas. First, roughly equal-sized parcellations are obtained. Second, these random parcellations are applied to individual subjects multiple times and a pseudo-bootstrap (PBS) of the network is obtained for statistical inferences. It was found that the variation of mean global network metrics from PBS sampling is smaller compared with inter-subject variation or within-subject variation between two diffusion MRI scans. Using the mean global network metrics from PBS sampling, the intra-class correlation is always higher than the average obtained from using a single random parcellation. As one application, the PBS method was tested on the Human Connectome Project resting state dataset to identify individuals across scan sessions based on the mean functional connectivity (FC)—a trivial network property that has little information about the connectivity between nodes. An accuracy rate of ∼90% was achieved by simply finding the maximum correlation of mean FC of PBS samples between two scan sessions.

KW - Connectomes

KW - Functional connectivity fingerprint

KW - Intra-class correlation coefficient

KW - Network analysis

KW - Pseudo-bootstrap

KW - Random parcellation

UR - http://www.scopus.com/inward/record.url?scp=85027837369&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85027837369&partnerID=8YFLogxK

U2 - 10.3389/fnhum.2017.00351

DO - 10.3389/fnhum.2017.00351

M3 - Article

VL - 11

JO - Frontiers in Human Neuroscience

JF - Frontiers in Human Neuroscience

SN - 1662-5161

M1 - 351

ER -