Contrast-to-noise ratios for indices of anisotropy obtained from diffusion MRI: a study with standard clinical...

Area of Science:

• Neuroimaging
• Diffusion MRI
• Biomedical Engineering

Background:

• Diffusion-weighted MRI has been utilized for 15 years to quantify diffusion anisotropy in healthy and pathological brain tissues.
• Accurate measurement of diffusion anisotropy is crucial for understanding brain microstructure and detecting abnormalities.

Purpose of the Study:

• To compare the performance of various diffusion anisotropy indices in differentiating tissue types.
• To evaluate the efficacy of different indices using both simulated and experimental diffusion MRI data.

Main Methods:

• Simulations were conducted using one-, two-, and three-fibre populations to model complex tissue microstructures.
• Performance of anisotropy indices was assessed based on their ability to differentiate between isotropic and anisotropic voxels.
• Comparison included indices derived from diffusion tensor models and model-free approaches, as well as different acquisition schemes.

Main Results:

• Indices from higher-rank tensors and model-free approaches effectively differentiate isotropic voxels from three orthogonal fibre populations.
• Geodesic anisotropy (GeoA), generalised anisotropy (GA), and scaled entropy (SE) demonstrated superior contrast-to-noise ratios compared to fractional anisotropy (FA) in simulations and for large anisotropy differences.
• Biological variability within brain regions can mask subtle differences between indices, limiting their effectiveness for small anisotropy variations.
• Utilizing multiple b-values in acquisition schemes appears to enhance contrast-to-noise ratios for traditional diffusion tensor model-derived indices.

Conclusions:

• Higher-rank tensor and model-free diffusion MRI indices offer improved differentiation capabilities for complex microstructural environments.
• While advanced indices show potential, biological scatter necessitates caution when interpreting small anisotropy differences in vivo.
• Optimizing acquisition schemes, such as using multiple b-values, can improve the performance of established diffusion tensor imaging metrics.