Related Experiment Video
Updated: Dec 22, 2025
09:57
How to Measure Cortical Folding from MR Images: a Step-by-Step Tutorial to Compute Local Gyrification Index
Published on: January 2, 2012
28.4K
Influence of Processing Pipeline on Cortical Thickness Measurement
Shahrzad Kharabian Masouleh1, Simon B Eickhoff1,2, Yashar Zeighami3
1Research Centre Jülich, Institute of Neuroscience and Medicine (INM-7: Brain and Behavior), 52425, Jülich, Germany.
Cerebral Cortex (New York, N.Y. : 1991)
|May 8, 2020
Summary
Neuroimaging analysis pipelines yield reliable gray matter (GM) thickness estimates within themselves. However, direct comparisons across different pipelines are cautioned due to inter-individual variability, especially in specific brain regions.
Area of Science:
- Neuroimaging and Computational Neuroscience
- Brain Structure and Morphology Analysis
Background:
- Replicability of neuroscientific findings, particularly gray matter (GM) morphological correlates, faces scrutiny.
- Variations in processing pipelines and their gray matter (GM) estimates may impact study reproducibility.
Purpose of the Study:
- To investigate and compare cortical thickness estimates from three widely used neuroimaging processing pipelines.
- To assess the reliability and comparability of cortical thickness measurements across different analytical approaches.
Main Methods:
- Analysis of cortical thickness estimates from three distinct, commonly utilized neuroimaging processing pipelines.
- Utilized two independent, large-scale cohorts for robust validation of findings.
- Evaluated both within-pipeline reliability and between-pipeline correlation of thickness estimates.
Main Results:
- High within-pipeline reliability for absolute cortical thickness estimates was observed.
- Comparable spatial patterns of cortical thickness estimates were found across all pipelines.
- Significant between-pipeline correlations in cortical thickness estimates were noted at the group level for most brain regions.
- Lower correlations and higher interindividual variability were found in para-limbic areas and the insula.
Conclusions:
- While within-pipeline estimates are reliable, in-vivo cortical thickness measurements are pipeline-dependent proxies.
- Comparisons of absolute cortical thickness should be restricted to data processed with the same pipeline and technique.
- Structural variations in regions like the insula and para-limbic areas require cautious interpretation due to lower reliability and higher variability.
Keywords:
in-vivo cortical thicknessinterindividual variabilityreliabilityreplicabilitysoftware comparison
