A crucial question regarding MRI using hyperpolarized gas is: will there be enough hyperpolarized gas in the organ of interest to produce useful images? This model attempts provide some of the information that can help answer that question.
Upon first consideration, it seems as though there is not enough
information to answer this question. However with further contemplation,
one quickly realizes that actually a lot is known about the kinetics
of inert gases and that the only hazy information is the T1's
of the hyperpolarized gas in various tissues. Thus, a reasonable
approach is to create a model, put in the best known values for
T1 and see if the numbers are even in the ballpark. As better
estimates of the T1's become available, they can be plugged into
the model. And that is what this web site provides; an opportunity
for you to run our model with your favorite kinetic parameters.
On inspiration, the hyperpolarized gas flows from the reservoir into the mouth and throat and in to the lungs. In the lungs it is picked up by the pulmonary blood and transported to the organ of interest. In that organ it diffuses both into and out of the tissue. We assume that by the time the hyperpolarized gas reaches the large veins, all of the hyperpolarization is gone. On expiration, the hyperpolarized gas flows from the lungs into the throat and mouth and leaves the system.
Quantitative results can be obtained by converting the model into a system of differential equations and solving them. Different gas delivery strategies can be modeled by altering the initial conditions. Our model implements three delivery strategies: continual breathing, hyperventilation followed by a breath-hold and hyperventilation followed by continual breathing. One delivery strategy that maybe useful for animal work is to continually blow a fixed concentration of HpGas directly into the animal's lungs. In our model, this case can be achieved by using the breath-hold strategy and setting the blood flow through the lungs to 0 and setting the T1 in lungs to a large value.
Our model is fairly general. The major limitation is that our model does not account for recirculating HpGas. In the human brain this should not produce significant errors. But in other species or other organs this may not be the case. To use the model for a specific organ, you will need to know the organs blood flow, partition coefficient, T1 and the time it takes for the blood to travel from the lungs to the organ. To use the model for a specific gas, you will need to know the solubility of that gas in blood, its T1's in gas, blood and tissue, and its partition coefficient. Note, the gas must freely diffuse between the blood and the organ of interest. In order to use this model in other species, we have made all of the physiological parameters adjustable by the user.
One application of our model that may be of interest, is the estimation of the concentration of stable gas in the brain. Although the concentration of HpGas reaches a peak quickly and is relatively small, the concentration of stable gas continues to rise for several minutes and is quite a bit larger than the concentration of HpGas. This is of interest for xenon because xenon is known to be a potent psychoactive substance and is an anesthetic at 71%. The stable gas concentration can be estimated by setting all the T1's in the model to their maximum value. The calculated concentrations will slightly underestimate the true concentrations because our model does not account for recirculating gas. Nevertheless, the model illustrates the important differences between the kinetics of stable gas and HpGas.
The details of our model have been published in the September
1997 issue of Journal of Magnetic Resonance Imaging.
Recently a similar model was published by Peled et al. (Magnetic
Resonance in Medicine, 1996, Vol. 36). You might want to check
this paper out.
If you have any questions please contact me at: martin@uthscsa.edu
Other Web Sites Related to Hyperpolarized Gas Magnetic Resonance:
APS Article on Hyperpolarized Xenon
University of Michigan Hyperpolarized Xenon MRI Research
Princeton University Hyperpolarized Gas MRI Research
An image of guinea pig lungs using HpHe MRI
Stella Modeling Software. If you are interested in creating your own model you might want to check out this software. We did not use this software for our model, but we have used it in the past and found it to be very good.
Optopower: A Source for Lasers. If you're thinking about building your own polarizer you might want to check this site out.
Combine Shirley Mclain and Noble Gases and what do you get? Click here to find out.
