Correlation tensor magnetic resonance imaging.

Area of Science:

• Magnetic Resonance Imaging
• Biophysics
• Neuroscience

Background:

• Diffusional Kurtosis Imaging (DKI) is a sensitive biomarker for tissue microstructure but lacks specificity.
• Existing methods to decouple kurtosis sources assume multiple Gaussian diffusion components, limiting their applicability.
• Intra-compartmental kurtosis, crucial for understanding restricted diffusion, is difficult to isolate.

Purpose of the Study:

• To develop a general framework for resolving kurtosis sources in diffusion MRI without Gaussian approximations.
• To introduce Correlation Tensor MRI (CTI) for explicitly decoupling isotropic and anisotropic kurtosis components.
• To derive an index sensitive to intra-compartmental kurtosis for improved microstructural analysis.

Main Methods:

• Developed Correlation Tensor MRI (CTI) using double diffusion encoding (DDE).
• CTI leverages displacement correlation tensors to decouple kurtosis sources.
• Applied CTI to ex vivo mouse brains (16.4 T) and in vivo rat brains (9.4 T).

Main Results:

• CTI successfully decoupled microscopic anisotropy from partial volume effects.
• Anisotropic and isotropic kurtosis components were accurately resolved.
• An intra-compartmental kurtosis index showed positive values in brain tissues.
• In vivo CTI results were consistent with ex vivo findings.

Conclusions:

• CTI provides a more general and accurate method for estimating kurtosis contributions in diffusion MRI.
• The technique can distinguish between different sources of non-Gaussian diffusion.
• CTI shows promise for in vivo studies, enhancing understanding of tissue microstructure in health and disease.
• Further research is needed to refine the intra-compartmental kurtosis index by addressing higher-order effects.