Related Experiment Videos
Sensitivity and specificity considerations for fMRI encoding, decoding, and mapping of auditory cortex at ultra-high
Michelle Moerel1, Federico De Martino2, Valentin G Kemper3
1Center for Magnetic Resonance Research, Department of Radiology, University of Minnesota, Minneapolis, USA; Maastricht Centre for Systems Biology, Maastricht University, Maastricht, The Netherlands; Department of Cognitive Neuroscience, Faculty of Psychology and Neuroscience, Maastricht University, Maastricht, The Netherlands; Maastricht Brain Imaging Center (MBIC), Maastricht, The Netherlands.
Ultra-high field functional MRI (fMRI) acquisition parameters influence spatial specificity. T2*-weighted imaging offers broad coverage for decoding, while T2-weighted imaging provides higher specificity for detailed mapping in auditory cortex research.
Area of Science:
- Neuroimaging
- Magnetic Resonance Imaging
- Auditory Neuroscience
Background:
- Ultra-high field functional MRI (fMRI) advances spatial resolution in studying neuronal activity.
- fMRI's blood-oxygen-level-dependent (BOLD) contrast relies on vascular signals, making spatial specificity dependent on vasculature characteristics.
- Acquisition parameters at 7 Tesla (7T) modulate the contribution of macro- and microvasculature to the fMRI signal.
Purpose of the Study:
- To investigate how fMRI measurement parameters affect spatial specificity in high-end analyses at 7T.
- To compare T2*-weighted (2D GE EPI) and T2-weighted (3D GRASE) acquisitions for encoding, decoding, and submillimeter mapping in the human auditory cortex.
Main Methods:
- Acquisition of T2*-weighted fMRI data using 2D gradient echo (GE) Echo Planar Imaging (EPI).
- Acquisition of predominantly T2-weighted fMRI data using 3D General Relativity Anisotropic Segmentation (GRASE).
- Comparison of decoding accuracy using two encoding models and mapping of voxel preferences (tonotopy) between the two acquisition types.
Main Results:
- T2*-weighted GE-EPI showed higher decoding accuracy due to greater spatial coverage and sensitivity, but exhibited cortical depth-dependent vascular biases.
- Submillimeter tonotopic maps revealed biases in frequency preference and selectivity with GE-EPI, whereas 3D GRASE maintained specificity.
- Large-scale tonotopic maps were similar, but GE-EPI offered better coverage and sensitivity.
Conclusions:
- Different fMRI acquisition strategies are optimal for distinct neuroimaging analyses.
- T2*-weighted acquisitions are advantageous for encoding/decoding analyses in auditory cortex research.
- T2-weighted acquisitions are recommended for high-specificity tonotopic mapping, crucial for detailed auditory cortex investigations.