Human cerebellar microstructure with 7T MRI: Spherical b-tensor encoding and super resolution reconstruction

Area of Science:

• Neuroimaging
• Magnetic Resonance Imaging (MRI)
• Cerebellar Neuroscience

Background:

• Cerebellar structure and function are less understood than the cerebrum.
• In vivo visualization of cerebellar microstructure is limited despite clinical relevance.
• The cerebellar granular layer, rich in neurons, is affected in various diseases.

Purpose of the Study:

• To implement and evaluate advanced diffusion MRI techniques at 7T for in vivo cerebellar microstructure imaging.
• To combine spherical b-tensor encoding, B1+ field shimming, and super-resolution reconstruction (SRR).
• To assess the method's ability to infer cell density and compare findings with existing data.

Main Methods:

• Utilized spherical b-tensor encoding with free waveforms at 7T.
• Incorporated B1+ field shimming and super-resolution reconstruction (SRR).
• Compared diffusion-weighted and normalized diffusion signals in cerebellar and occipital grey/white matter, and 30 cerebellar ROIs.

Main Results:

• Successfully implemented spherical tensor encoding and SRR at 7T for human cerebellum imaging.
• Normalized diffusion signal, related to cell density, was increased in the cerebellar cortex vs. occipital cortex.
• Cerebellar normalized diffusion signal patterns correlated with known granular layer thickness variations across lobules.

Conclusions:

• Advanced diffusion MRI at 7T can effectively image human cerebellar microstructure in vivo.
• The method provides insights into cerebellar cell density, consistent with post-mortem studies.
• Findings support the clinical relevance of high-resolution cerebellar imaging for understanding neurological conditions.