MRI question for our Heavy Weights

[Tsu]

OMG! OH... OH...! OMG!

*Tsunami dancing all over the place, on the couch, on table, on the piano bench, on the piano...*

I followed and understood everything you guys said! Even pervect's equation! WOOHOOOOOO! I'm a PHYSICIST!

imabug - I argree with you on the image recon algroithms. It's the same in CT and CR. But its imperative that we know what each and every algorithm does, so that we can apply them in the proper situations. I was able to show my Rad a basilar skull fx with associated small subdural hematoma (that he had missed) by knowing how to use the algorithms and manipulate the data on my new scanner. That same week, I demonstrated a tiny area of bleeding (secondary to trauma) in the brain of a young man. It changed his course of treatment.

 

[motai]

motai - Our PET scanners in Oregon are basically traveling minstrel shows. They are doing what CT and MR did in their infancy and putting the scanners on giant trucks and covering numerous hospitals all over the state. I don't even think they have a permanent PET scanner up in Portland. The place where I was working in Portland (a VERY large health system) had just set up a mobile PET pads at their hospitals. Same goes with the hospital where I now work. I had a blast doing mobile CT back in the '80's but I'm way too old to be a traveling tech again. I now live on a little farm with numerous critters that takes up WAY more time and energy than an apartment with no critters. I have made up my own version of PET SCANNING! LIVE ON A FARM!

*pulls out PETnet pamphlet and looks at a map*... there is a PETnet facility in Portland, and the only other one in that geographical region is in Seattle. There are a few facilities in California, almost none in the box states (w/exception to Denver), some scattered through the mid-west, and lots in the Eastern half of the United States. Their center of operations is in Knoxville and their customer service provider is in Orlando.

http://www.petnetpharmaceutical.com

That is... unless the PETnet facility is your farm w/critters...

 

[pervect]

The precession angle (or more commonly called flip angle) is determined by how long you apply the RF pulse to the body.

OK, there's something I must be fundamentally confused about here, it's been too long since I've used my quantum mechanics.

They way I recall quantum spin, the spin state could be completely described as a superposition of |up> and |down> spins. Pick an axis, any axis, and the total spin can be described as a superpositon of up & down spins along that axis. The Stern-Gerlach experiment is the cannonical example - a beam of silver ions that splits into two parts when passing through a spatially varying magnetic field.

So I always imagined the spins as either being aligned or anti-aligned, i.e. either |up> or |down>, but the descriptions I see on the WWW are of precession.

 

[Tsu]

*pulls out PETnet pamphlet and looks at a map*... there is a PETnet facility in Portland, and the only other one in that geographical region is in Seattle. There are a few facilities in California, almost none in the box states (w/exception to Denver), some scattered through the mid-west, and lots in the Eastern half of the United States. Their center of operations is in Knoxville and their customer service provider is in Orlando.

http://www.petnetpharmaceutical.com

That is... unless the PETnet facility is your farm w/critters...

It's just that getting into PET will require a lot of travel and being away from home. I just don't want to do that anymore. I liked it a lot when I did it years ago, but now I'm a homebody! I like to hang with my critters and dig in my gardens.

 

[Tsu]

I think this may answer your question. I'm copying out of a MR registry study manual.

"When the protons' net magnetization is aligned with the main magnetic field, there is no measurable signal and we have no information to make an image or to determine anything unique about the appearance of normal and abnormal tissue. To learn about the tissue, we must disturb the proton's equilibrium and move their net magnetization out of alignment with the main magnetic field. (This is done with the RF signals) Once we stop the applied distubance, the magnetization will move back, or relax, toward its equilibrium position through two distinct relaxation processes. It is through the relaxation processes that the tissues can be distinguished in an image. The precessional motion of the proton's magnetization is very important because it occurs at a very specific rate or frequency. This is expressed by the Larmor Equation." (which all of you already know about, right? )

Please let me know if this cleared up your confusion or just caused more. (I'm famous for the latter... )

 

[imabug]

OK, there's something I must be fundamentally confused about here, it's been too long since I've used my quantum mechanics.

They way I recall quantum spin, the spin state could be completely described as a superposition of |up> and |down> spins. Pick an axis, any axis, and the total spin can be described as a superpositon of up & down spins along that axis. The Stern-Gerlach experiment is the cannonical example - a beam of silver ions that splits into two parts when passing through a spatially varying magnetic field.

So I always imagined the spins as either being aligned or anti-aligned, i.e. either |up> or |down>, but the descriptions I see on the WWW are of precession.

Most explanations of MRI represent spin using the spinning top model (not really accurate, but simplistic enough for radiologists to understand). MRI can only be done with atoms having a net magnetic moment. It's the magnetic moments that get 'flipped' by the RF pulse (actually it's just getting rotated, no spin flipping taking place at all). As the magnetic moments realign themselves with the main field, that RF energy is radiated out again. The common layman's representation of this is a bar magnet in an external magnetic field. If you rotate the magnet so that it isn't aligned with the external field, it will want to realign itself with the external field.

 

[pervect]

OK, I found a website that explains why precession is equivalent to a superposition of up & down states - I think I see what's going on now. The bar magnet explanation _is_ a lot easier than the formal quantum mechanical one!

http://www.phys.cwru.edu/courses/p431/notes-2003/node30.html
and node31, too.

Also, the explanation that the protons do naturally line up with the magnetic field, and you have to disturb them in order to do measurements was helpful - thanks, Tsunami.