Experimentally and computationally fast method for estimation of a mean kurtosis

Area of Science:

• Neuroimaging
• Biomedical Engineering
• Radiology

Background:

• Mean kurtosis (MK) derived from magnetic resonance diffusion imaging is a sensitive marker for tissue pathology.
• Current MK estimation methods are hampered by lengthy acquisition and postprocessing times, limiting clinical utility.

Purpose of the Study:

• To introduce and evaluate a novel, rapid protocol for estimating MK.
• To address the time constraints associated with conventional MK estimation.

Main Methods:

• A new MK estimation method using a rapid protocol with 13 diffusion-weighted images.
• Linear combination of log diffusion signals, avoiding nonlinear optimization.
• Evaluation on ex vivo rat and in vivo human brain datasets, comparing with standard diffusion kurtosis imaging (DKI).

Main Results:

• The new MK metric demonstrates contrast similar to conventional MK.
• The protocol achieves full human brain coverage in under 1 minute with seconds of postprocessing.
• Scan-rescan reproducibility is comparable to standard MK estimation.

Conclusions:

• The proposed framework provides a robust and rapid method for MK estimation.
• The protocol is easily adaptable to commercial MRI scanners with minimal modifications.
• This rapid MK estimation is feasible for widespread clinical application.