Lecture 18= Diffusion MRI Flashcards
What is a diffusion MRI based on?
- Based on the diffusion/ movement of water
- This movement is restricted more in some areas than others (moves along axon much better than across axons) therefore, MRI can give information about the direction of axons in the white matter
Does a diffusion MRI give information about the microstructure of the brain? What is the limitation to this?
-Yes, because it tells you about the direction of axons in WM however, this is average across a whole voxel (3D pixel)
In diffusion MRI why are lots of images required?
Because each image is only sensitive to one direction
How do images gained from structural and diffusion MRIs differ in look and time to take the scan?
- Structural MRI takes ~5 minutes to acquire one image while Diffusion MRI takes lots of fast images (one every 1-3 seconds) for a total of ~5 minutes (100-300 images+)
- To help with fast scanning, diffusion MRI has lower resolution (1-3mm) than structural MRI (<1mm) and this results in the image being a bit more fuzzy as opposed to the clear/ crisp features shown in a structural MRI
What is the corpus callosum?
- Bundle of axon fibers (white matter)
- Serves as a connection between the two hemispheres of the brain (highway)
How does the physics of a diffusion MRI work?
- The gradient coils create a magnetic field that can change in any given direction
- Moving spins (hydrogen and water molecules: attracted to magnet) in the gradient direction diffuses spins along that axis and reduces the signal (because signaling is very fast when it aligns)
What are tensors? What can they tell us?
Surrogate measures of microstructure:
- Average direction of axons in voxel
- Can indicate axonal integrity (e.g. mean diffusivity). If the tensors are fatter in shape then they have a higher mean diffusivity i.e. the diffusion is not as strong in one direction (along the axon) there is diffusion in every direction.
What are the two types of white matter tract tracing?
- Tractography traces connections via local directions
- Probabilistic tractography
Patients with what condition have greater mean diffusivity?
Trigeminal neuralgia (Severe Headaches)
As opposed to a t-test what is an alternative stat measure that can be applied to a research question surrounding MRI’s? How is this approach different?
Regression. Looking at how the two factors are related rather than if there is a difference (T-test)
What are the three things that need to happen when looking at a diffusive MRI picture to be able to start answering stats questions?
- Go from image to usable data
- Fit the tensor model, then segment the area you want (in this case= corpus callosum), Average the mean diffusivity measures to get 1 value for each individual)
How can a regression equation be formed for the question: How does mean diffusivity relate to age?
MD= intercept + (slope x age)
What does the P value for a regression show us?
How good of a line our regression model is to fit the data we have obtained. Less than 0.05 is good (significant)
What are some limitations of a diffusion MRI?
- Does not measure axon size/density directly
- Does not measure single fibres (only average groups)
- More difficult to deal with crossing/kissing fibres
- More difficult to do in pulsatile regions (e.g. brainstem)
- More restricted by scanner hardware
- Sensitive to fast imaging artefacts (hard to do very front of the brain)
What are complimentary techniques to a diffusion MRI? How does it compare as a technique?
- Tracer studies (individual fibres, need post mortem/ dead cells)
- Histology (myelin/axon dimensions/glia, only show limited area and again need to be post mortem)
-Diffusion MRI’s can show us axonal direction over the whole brain and can be done on living tissues however, does have its own constraints (in previous card).
