Week 06: Diffusion Tensor Imaging (DTI)
Methods in Human Neuroimaging — NBL 425/625While functional MRI maps the brain's cortical processors, Diffusion Tensor Imaging maps the structural wiring connecting them. This week covers Brownian motion, Stejskal-Tanner diffusion weighting gradients, the diffusion tensor ellipsoid, Fractional Anisotropy (FA), Mean Diffusivity (MD), and tractography.
Interactive Demos & Activities
-
DTI Tensor & Tractography Explorer
Shape the diffusion tensor by its eigenvalues and watch the six Stejskal-Tanner gradient measurements it would produce, then seed deterministic streamlines into a synthetic corpus-callosum / corticospinal crossing. Tune the FA stopping threshold, turning-angle limit, step size, and Euler vs. Runge-Kutta integration to see tracts terminate at the intersection, and to see them jump bundles and fabricate a connection when the angle limit is relaxed.
-
Interactive Fractional Anisotropy (FA) Simulator
Adjust the FA slider from 0.0 to 0.95+ to observe how water diffusion transitions from an isotropic sphere (free fluid in CSF) to a highly constrained ellipsoid parallel to densely packed, myelinated axonal fiber bundles.
-
129Xe Microstructural Diffusion Primer
Explore how restricted Brownian motion is quantified using Apparent Diffusion Coefficient (ADC) mapping to probe microscopic anatomical barriers.
Topics Covered
- Brownian Motion & Restricted Diffusion: Einstein's diffusion equation, isotropic diffusion in CSF, and anisotropic restriction by axonal membranes and myelin sheaths.
- The Diffusion Tensor Matrix: Applying diffusion gradients across $\ge 6$ non-collinear directions to calculate the $3 \times 3$ symmetric diffusion tensor and its principal eigenvalues ($\lambda_1, \lambda_2, \lambda_3$).
- Tractography & Structural Connectomics: Deterministic streamline algorithms (Euler/Runge-Kutta), probabilistic tractography, fiber crossing limitations, and whole-brain structural connectivity graphs.