Kicking the study part off in "The HSC with AB" is fittingly, my favourite subject, Physics. While I have complained a small amount about doing Medical Physics instead of From Quanta to Quarks, I do understand that there are a lot of people (read: Asians) who want to be doctors, so they really don't have that much of a choice. Please note that I like to have a lot of paragraphs, to avoid a wall of text that a) won't look too appealing and b) will not help divide potentially long-winded explanations. Bite-sized chunks is what I'm aiming for.
Today I'm going to be looking at MRI, which is quite possibly one of the most complex things we've done in Physics so far, but I think seems a lot harder than it actually is. That's not to say it's not difficult, but it's not rocket surgery.
MRI is, in short, measuring the magnetic properties of hydrogen nuclei (or, protons) in water molecules in a patient. As different parts of the body have different water contents, each part of the body has different concentrations of hydrogen nuclei. This in turn means each part of the body has different magnetic properties, and as such, MRI can produce a three-dimensional image of the body. It's a little more complicated than that, obviously, but that's the nuts and bolts of it. If you take this much out of this post, I consider my work done (no, not really. But I figure it's a good introduction to a complex subject.).
I'm going to be using a lot of terminology in this, so allow me to preface with some definitions. Firstly, when I say antiparallel, I don't mean perpendicular. I mean parallel in the opposite direction. Secondly, precession is the rotation of an axis of a rotating object. While that may sound confusing, imagine a top, or, better yet, look to the right. That's precession. Lastly, a voxel is like a three-dimensional pixel. A cube instead of a square (if you play Minecraft, imagine every cube is a voxel.) OK, with that out of the way, let's jump into MRI.
Firstly, the patient lies inside the MRI machine, with an intense magnetic field parallel to the body (called the longitudinal direction, or the Z-axis). The protons in the water molecules will tend to align their spins with the field (imagine a top spinning. It tends to align its spin with gravity. A similar thing is happening here.). Most of the protons will align parallel, but some will align antiparallel.
Radio waves are then passed through the patient. This causes two things. One, it causes an equal number of protons to align parallel and antiparallel to the field. Two, the precession of the protons synchronises.
That might seem confusing (and if it doesn't, congratulations), but I'll try to explain better. I want you to stand up and move your arms in circles, almost like you are doing the butterfly (the swimming style that everybody hates). Your arms are now the protons. There is, as you can see, an equal number of protons parallel and antiparallel to the field (one arm each). Also, the precession of the protons is synchronised (your arms are moving together at the same rate).
OK, back to our MRI machine, and back to our protons. These synchronised, precessing protons produce a net rotating magnetic field. This is entirely in the XY plane (if you like, imagine a string connecting your hands and rotate your arms). This produces a changing flux in the detecting coils which, if we know our Motors and Generators, produces a current.
Now, in our MRI machine, the radio waves stop. This reverses what happened when we applied our radio pulse earlier. Firstly, the protons with antiparallel spins revert back to having parallel spins over time, causing a longitudinal component to emerge. The time it takes for this component to emerge is called T1. Secondly, the precessing protons desynchronise, making the rotating magnetic field in the XY plane disappear. The time it takes for this to happen is called T2. These will become much more important later, but for now, we need to look at one more part of the MRI machine: the gradient coils.
These gradient coils simply produce a weak magnetic field which slightly alters the strong magnetic field of the MRI machine from point-to-point inside the patient. This means that at every point in the patient, the magnetic behaviour of the protons is different. Now here's the good part: this changes the T1 and T2 values in each voxel (if you're confused, look back at the definitions). The detectors can measure these differences and so can differentiate between the properties of each voxel.
What does this mean? Every mm3 of space has slightly different properties which, when fed into the computer, can be calculated. This produces a high-resolution, three-dimensional image of the inside of a patient.
To summarise: a powerful magnetic field, combined with a pulse of radio waves, generates a rotating magnetic field. When a weaker, gradient field is applied, it slightly alters the rotating magnetic field from point-to-point, which can be detected and processed to produce a detailed 3D image.
This may seem very confusing, and you may not get it all in the one go. It introduces a lot of new ideas, and not simple ones either. I would suggest taking a break, letting your mind take in some of the info, and coming back when you're ready. There's not going to be a test tomorrow (unless you're reading this November 3, in which case, good luck!).
For those weirdos who need impacts on society, MRI is very safe since it doesn't use dangerous ionising radiation like X-rays. It is expensive though, since it uses superconductors and liquid helium.
OK, I think this about wraps up MRI. If I've missed anything important, or if I've got something wrong, or if you've found an easier way to explain something, or anything really, then please let me know in the comments below.
With MRI ticked off the list,
AB
hey AB, instead of saying that MRI measures the magnetic properties of hydrogen nuclei, maybe say that they use the magnetic properties to measure the water content
ReplyDeleteBy "weirdos who need impact on society" do you mean the board of studies?
ReplyDeleteGradient coils are used to differentiate between different voxels (the pun? you're welcome.)
But yeah, very good, I like it, looking forward to the next post =)
Ebony (can't be bothered to sign in)
nice background
ReplyDeleteNice. Very concise.
ReplyDeleteEric.