NMR and MRI Theory: Understanding Relaxivity for Contrast in Imaging

[Coto]

NMR and MRI -- questions!

I'm currently doing an experiment on nuclear magnetic resonance for an undergrad physics lab. Our experiment was based around determining the effects of proton concentration on spin-lattice and spin-spin relaxation times. We used two separate sets of ionic solutions (cupric sulfate solutions and ferric nitrate solutions.)

To analyze the results I'll be looking at T1 and T2 vs. concentration level as well as the supposed 1/T1 1/T2 relaxivity vs concentration. My problem is I can find little to no theory whatsoever that describes to me in detail the importance of relaxivity. I'm aware that relaxivity affects contrast in MRI images and to what extent it can do so, however I'm not sure about what makes a substance good for T2 imaging versus good for T1 imaging?

I guess in short, other than that main question at the end, I'm looking for sources about MRI relaxivity theory. Any help would be greatly appreciated.

EDIT: Sorry about posting this in the main forum. Since I was just looking for sources on MRI theory, I thought it'd be just as fine to post in the general forum.

 

[Coto]

Found the answer to my own question.. may as well post it in case someone else is needing the same help one day.

MRI imaging uses voxels that are governed by an intensity equation based off of T1 and T2 relaxation times. For 2d images, the pixel intensity is related as:

I = \kappa \rho(x,y) (1-exp({\frac{-\tau_R}{T_1}})) exp({\frac{-\tau_e}{T2}}})

Where \tau_R and \tau_e are the repetition time and echo pulse time respectively, \rho(x,y) is the proton density w.r.t 2 dimensions, and \kappa is a proportionality (normalization) constant.

As can be seen shortening T1 will increase pixel intensity, where as shortening T2 will decrease pixel intensity.. this naturally leads to T1 and T2-weighted imaging in MRI application.

Gyah, I hope that helps someone someday. Ah. For sources, try and find any modern book on MRI technology.. NMR books generally don't mention this stuff.