Q-space trajectory imaging for multidimensional diffusion MRI of the human brain

Area of Science:

• Biomedical Imaging
• Diffusion MRI
• Neuroscience

Background:

• Traditional diffusion MRI methods like pulsed field gradient sequences probe single points in q-space.
• Existing techniques average diffusion information within a voxel, limiting detailed microstructure analysis.

Purpose of the Study:

• Introduce a new diffusion MRI framework, q-space trajectory imaging (QTI), for advanced microstructure imaging and modeling.
• Develop a diffusion tensor distribution (DTD) model to leverage QTI's enhanced data for more comprehensive tissue analysis.

Main Methods:

• Propose q-space trajectory encoding using time-varying gradients to probe trajectories in q-space.
• Introduce the diffusion tensor distribution (DTD) model for estimating distributions of diffusion tensors.
• Define the b-tensor as a tensor-valued extension of the b-value, enabling higher-order tensor analysis.

Main Results:

• QTI enables microstructure modeling beyond the capabilities of traditional pulsed gradient encoding.
• The b-tensor allows estimation of the mean and covariance of the DTD model using second and fourth-order tensors.
• Rotationally invariant scalar quantities for size, shape, and orientation coherence were derived.
• A pilot study showed significant differences in 9 out of 14 QTI-derived parameters between healthy controls and schizophrenia patients.

Conclusions:

• QTI provides a powerful new paradigm for studying complex tissue architecture by modeling diffusion tensor distributions.
• This framework enhances the discrimination of microenvironmental features like size, shape, and orientation coherence.
• QTI demonstrates potential for clinical applications, as evidenced by its ability to detect group differences in schizophrenia patients.