Related Experiment Video
Updated: Apr 19, 2026
05:55
Modeling the Functional Network for Spatial Navigation in the Human Brain
Published on: October 13, 2023
1.7K
Reproducibility of graph-theoretic brain network metrics: a systematic review
Thomas Welton1, Daniel A Kent, Dorothee P Auer
1Sir Peter Mansfield Imaging Centre, University of Nottingham , Nottingham, United Kingdom .
Brain Connectivity
|December 11, 2014
Summary
This review found that brain network metrics generally show good test-retest reliability in healthy subjects. Structural brain networks may be more reproducible than functional ones over time.
Area of Science:
- Neuroscience
- Network Science
- Biostatistics
Background:
- Graph-theoretic brain network metrics are increasingly used to analyze brain structure and function.
- Assessing the reproducibility of these metrics is crucial for their reliable application in research and clinical settings.
Purpose of the Study:
- To systematically review and assess the test-retest reliability of graph-theoretic brain network metrics.
- To identify factors influencing the reproducibility of these neuroimaging metrics.
Main Methods:
- Systematic review of primary research studies on test-retest reliability in healthy human subjects.
- Inclusion criteria: quantification of reliability using intraclass correlation coefficient (ICC) or coefficient of variance.
- Databases searched: MEDLINE, Web of Knowledge, Google Scholar, OpenGrey (up to Feb 2014). Risk of bias assessed using 10 criteria.
Main Results:
- Twenty-three studies (n=499 subjects) were included, with retest intervals ranging from <1 hour to >1 year.
- Intraclass correlation coefficients (ICCs) reached fair, good, or excellent ranges in 1, 5, and 16 studies, respectively.
- Overall reproducibility was good; structural network metrics demonstrated higher reliability than functional metrics in most cases.
Conclusions:
- Graph-theoretic brain network metrics exhibit good overall reproducibility.
- Structural brain networks appear to be more reliable over time compared to functional networks.
- Methodological factors significantly impact the reproducibility of brain network metrics.
