Talk:Magnetic resonance imaging

neuroimaging

I added this red link (as an uncreated article - is it the best term?) - I won't be creating it though. If anyone else wants to... --Matt Lewis 00:44, 31 March 2008 (CDT)

The piece of the pictue that I miss

Your section on how various weightings and gradients return information of different brighness or color is very helpful. What still puzzles me about MRI, vs. CT ad SPECT, is that in the latter, there is a narrow collimated beam that is "spread" over a limited angular range--i.e., tomography. I have at least a conceptual grasp of how the image is built from the set of tomographic images.

As far as I know, magnetic and RF fields cannot be collimated to anywhere close to the precision of collimated ionizing ratioactive. The thing I don't yet grasp in MRI is how the image is formed form the individual images. Can you help define that, preferably comparing it with SPECT and CT? Howard C. Berkowitz 03:46, 29 July 2008 (CDT)

By using gradient excitation fields in 2 or more directions, you can effectively wipe out all signals except for those in a small region, say a 1 mm squared point where the X-axis and Y-axis radio frequencies meet. Then, by collecting data for many small regions, overlapped a bit, you can build up the image. The greatest natural contrast occurs between aqueous and lipidic regions, because water moves very fast and large lipids move very slowly, contributing to very different relaxation rates. Contrasts can be artificially increased by using specially labeled compounds, which for example, might bind to certain receptors, or nerves, and so on. David E. Volk 23:11, 29 July 2008 (CDT)