Spectroscopic Imaging (MRI is so cool ...)

• know that the chemical shift changes the Larmor Frequency for $^1$H protons in different chemicals. Therefore, the spins are going to be behaving differently under same imaging conditions.
  $$\begin{eqnarray*} s(\vec{k},t) &=& \int\, d^{3}r\, d \sigma\, \rho(\vec{r},\sigma) e^{-i2\pi( \vec{k}\cdot \vec{r} - \sigma f_{0}t)} \end{eqnarray*}$$
  
  
  
  
  

• can we make use of this information to image individual chemical species with $^1$H protons??? SURE!!!, NO PROBLEMO!!! at least, in principle

• Look at 2D imaging equation for a SE experiment with the effect of chemical shift added, and relaxation neglected
  $$\begin{eqnarray*} s(k_x,k_y,t') = \int d \sigma\,\int dx\,dy\,\rho (x,y,\sigma)e^{-i2\pi (k_{x}x+k_{y}y-\sigma f_{0}t')} \end{eqnarray*}$$
  
  
  
  
  

• Let's see what the associated sequence diagram would look like
  
  

• since all spatial (phase) encoding is done before sampling all of the information we are now picking up by sampling at different times is related to the chemical shift of the species relative to free water, for example

• Unfortunately, phase encoding in an extra dimension is going to cost (big time) in the overall time to complete the experiment
  $$\begin{eqnarray*} T_{acq} = N_{x}N_{y} T_{R} \end{eqnarray*}$$
  
  
  
  
  
  (IN-CLASS QUESTION)

• what kind of `` images'' do you expect to get out of a CSI experiment???
  
  

  
  
  
  
  
  
  
  

• Why do spectroscopy??? Most imaging is based upon anatomy, or physiology as it relates to blood flow. That just scratches the surface of physiological information that doctors want. Beyond imaging, you cut bits and pieces of people out, and send them to labs. Spectroscopy holds the promise that you could do some of this ``lab'' work, in house, so to speak.

• MRI problems with spectroscopy include dependence on perfect background field, small signal, long acquisition times. Other practical problems include lack of overwhelming data correlating MRS to physiological results, fear of data overflow.