Laminar analysis of 7T BOLD using an imposed spatial activation pattern in human V1

Area of Science:

• Neuroimaging
• Functional Magnetic Resonance Imaging (fMRI)

Background:

• High-resolution fMRI is limited by hemodynamic signal spread.
• Large cortical surface vessels bias the Blood-Oxygen-Level-Dependent (BOLD) signal away from neuronal activity.
• Spatial accuracy is affected by vascular distribution and measurement parameters.

Purpose of the Study:

• To introduce a laminar surface-based analysis method.
• To investigate the relationship between spatial localization and activation strength across cortical laminae.
• To assess the impact of avoiding superficial cortical layers on spatial resolution and signal strength.

Main Methods:

• Acquired 1mm isotropic, single-shot Echo-Planar Imaging (EPI) at 7 Tesla.
• Sampled the BOLD signal from superficial, middle, and deep cortical laminae.
• Utilized a boundary-based registration algorithm for aligning EPI data to surface reconstructions.

Main Results:

• Highly-accelerated EPI limited image distortions, enabling accurate registration.
• Spatial spread of the BOLD response was analyzed as a function of cortical depth.
• Sampling near the pial surface yielded the highest signal but also the most spatial error.

Conclusions:

• Avoiding superficial cortical laminae improved spatial localization by approximately 40%.
• This improvement came at the cost of a 36% decrease in z-statistic (signal-to-noise ratio).
• Anatomically-informed spatial sampling, avoiding large pial vessels, is crucial for optimal fMRI spatial resolution.