Related Experiment Videos
Identifying confounds to increase specificity during a "no task condition". Evidence for hippocampal connectivity
S A R B Rombouts1, C J Stam, J P A Kuijer
1Department of Physics and Medical Technology, VU Medical Center, Amsterdam, The Netherlands. sarb.rombouts@vumc.nl
Neuroimage
|October 22, 2003
Summary
This study used multiple linear regression to distinguish true hippocampal connectivity from noise in functional MRI data. The method successfully identified brain connectivity patterns related to the hippocampus during resting states.
Area of Science:
- Neuroimaging
- Cognitive Neuroscience
- Biomedical Engineering
Background:
- Functional MRI (fMRI) is used to study brain connectivity during resting states.
- Physiologic noise can create spurious connectivity patterns, confounding resting-state fMRI analyses.
- Distinguishing true neural signals from noise is crucial for accurate connectivity mapping.
Purpose of the Study:
- To apply multiple linear regression to identify and separate hippocampal connectivity from spurious signals in resting-state fMRI.
- To differentiate connectivity related to the blood oxygen level dependent (BOLD) signal from that caused by confounding factors like physiologic noise.
Main Methods:
- Utilized multiple linear regression analysis incorporating hippocampal time courses as regressors of interest.
- Included respiratory signals and cerebrospinal fluid data as regressors of no interest to model confounding factors.
- Applied the method across varying sampling rates (high/low) and spatial resolutions in five healthy subjects.
Main Results:
- Regressors of no interest demonstrated distinct connectivity patterns separate from hippocampal activity.
- Hippocampal regressors revealed significant connectivity between the left and right hippocampus.
- The analysis successfully generated separate connectivity maps for BOLD-signal-driven and spurious signals.
Conclusions:
- Multiple linear regression is an effective method for analyzing hippocampal connectivity in resting-state fMRI.
- The approach can successfully distinguish genuine hippocampal functional connectivity from noise-induced patterns.
- This technique enhances the reliability of fMRI studies investigating resting-state brain networks.