Towards higher sensitivity and stability of axon diameter estimation with diffusion-weighted MRI

Area of Science:

• Neuroimaging
• Biophysics
• Materials Science

Background:

• Diffusion-weighted magnetic resonance imaging (dMRI) is crucial for non-invasive in vivo axon morphometry.
• The ActiveAx technique models white matter to determine axon diameter and density.

Purpose of the Study:

• To investigate how gradient strength and model fitting routines impact the sensitivity of ActiveAx for small axon diameter estimation.
• To validate dMRI-derived axon diameter indices against electron microscopy (EM) in mouse brains.

Main Methods:

• Acquired ex vivo diffusion-weighted MRI data of mouse brains using 16.4-T MRI with high (300 mT/m) and ultra-high (1350 mT/m) gradient strengths.
• Employed and evaluated a dictionary-based fitting routine.
• Validated estimated axon diameter indices of the corpus callosum using electron microscopy.

Main Results:

• Higher gradient strengths resulted in axon diameter indices closer to EM measurements.
• Despite improvements, estimated axon diameter indices (~1.8 µm) were higher than EM measurements (~1.2 µm).
• Pulsed gradient spin echo (PGSE) sequence limitations and axonal dispersion were identified as factors affecting sensitivity.

Conclusions:

• Gradient strength significantly influences the accuracy of dMRI-based axon diameter estimation.
• Current dMRI techniques, including ActiveAx, have limitations in accurately measuring small axon diameters.
• Optimizing acquisition protocols, tissue models, and fitting routines is essential for advancing microstructural diffusion imaging.