Unfortunately the utility of MRIs (for human diagnostic purposes) does not scale linearly with power at least on the lower end. You need a certain resolution to reach a level of confidence to make a diagnosis and act upon it.

When radiologists or neurologists who are used to looking at slices from 3T machines have to look at lower quality images you can hear a few grumbles about how it would increase their diagnostic certainty about what they are looking at if they had just another slice or 2.

Still a very cool project though!