Investigating apparent differences between standard DKI and axisymmetric DKI and its consequences for biophysical...

Area of Science:

• Diffusion MRI
• Biophysical modeling
• Neuroimaging

Background:

• Diffusion kurtosis imaging (DKI) provides insights into microstructural tissue properties.
• Standard DKI and axisymmetric DKI (AxDKI) are two models for analyzing diffusion MRI data.
• Understanding their differences is crucial for accurate biophysical parameter estimation.

Purpose of the Study:

• To compare standard DKI and AxDKI models.
• To assess the impact of acquisition protocols on parameter estimation.
• To investigate the influence of fiber complexity on DKI parameter differences.

Main Methods:

• Simulated synthetic diffusion MRI data (noise-free and noisy) using standard and fast acquisition protocols.
• Estimated baseline differences between standard DKI and AxDKI models.
• Evaluated the effect of fiber complexity, acquisition protocols, and denoising on parameter precision.

Main Results:

• Significant baseline differences were observed for kurtosis parameters and WMTI-Watson model parameters.
• Fiber complexity, assessed via masks, reduced the number of voxels with large differences.
• The fast "199" protocol reduced precision in noisy data, but adaptive denoising mitigated this.

Conclusions:

• Axisymmetric DKI yields comparable diffusivity and mean kurtosis estimates to standard DKI under a standard protocol.
• Fiber complexity is a key factor driving differences between the DKI models.
• Adaptive denoising can counteract precision loss associated with faster acquisition protocols in noisy conditions.