14 fMRI contrast

We’ll open by talking about deoxyhemoglobin, and how the unpaired electrons on blood cells in veins perturb the magnet field. Then we’ll move on to talking about the most common (T2*-weighted, gradient echo) BOLD fMRI. Finally we’ll talk about other options for doing fMRI that have better spatial specificity (because the signal isn’t dominated by large veins) but lower contrast-to-noise ratio.

Probably 99% of fMRI experiments are done with GE EPI — it has low SAR and is easy to acquire. But if you need signal in regions where bone-tissue interface produce strong field perturbations, then you’ll want to do SE BOLD. Or, if you want to eliminate the signal from large veins and analyze just the small veins that blur the neuronal signal less, you’ll want to do SE BOLD. But if you do SE BOLD, you need to be prepared for higher SAR (because there’s a 180 degree re-focusing pulse), more variable sensitivity (because your refocusing pulse probably isn’t perfect everywhere), and a dramatic reduction in contrast to noise ratio. That’s OK — that huge BOLD signal was coming from the large veins you don’t want to see if you’re looking for sub-millimeter features in the patterns of neuronal responses.

Finally, here’s a short list of other things you can do if you have a really big magnet and care about spatial specificity in your signal: GRASE, VASO, CBF are the most popular.