MRI does not use ionizing radiation as does CT and conventional radiography. The MR machine is a very strong magnet, in the order of 150 times that of a conventional refrigerator magnet. With clinical imaging, we are interested in what happens to the hydrogen atoms (hydrogen atom is just one proton) in the region of interest. The magnetic field causes a small but significant number of the hydrogen protons in our body to realign with the main magnetic field, because the hydrogen protons themselves act like small magnets and respond to the main magnetic field. The majority of the hydrogen protons align with the magnet field, but some orientate in a higher energy level against the magnetic field. During imaging, a precise amount of energy is deposited into the patient in the form of a radiofrequency (RF) pulse (at 1.5 T this is at a frequency in the order of 64 MHz). Some protons absorb this energy and change from the low energy state to the higher energy state, aligning against the magnetic field. When the radiofrequency pulse is turned off, the hydrogen protons ‘excited’ by the RF pulse, give up that energy in the form of an RF signal, - they themselves become small radiofrequency transmitters. This RF signal, albeit very weak, is detected by coils around the patient and from this an image is generated. The way in which the hydrogen protons ‘give up their energy’ depends on the environment in which the proton resides. Protons in fat behave differently than protons in fluid or ligaments and this is the basis of image contrast. This is what enables us to see structures that are not visible on conventional radiographs.
The RF transmitted into the patient and emitted by the patient during imaging depends on the field strength, but is not dissimilar to that used in cell phones and radios. The MRI suite must be shielded from extraneous RF pollution from cell phones and radios and this is usually achieved by surrounding the magnet with RF shielding – usually copper or such like (faraday cage)
An image is basically a picture of a slice of tissue, one could liken it to a sliced loaf of bread. We simply lay the slices of bread out and look at each image to assess for abnormalities. We have options of buying bread sliced thinly for sandwiches or thicker for toast – there is more bread in a thicker slice and with MRI, thicker slices mean more hydrogen protons that can potentially contribute to the image. This result is in a better looking image. However, with thick slices, very small structures are often not visible, as the signal the abnormality produces is only a small fraction of the signal from the volume imaged, (voxel). Generally speaking a horse limb is pretty small compared to say a human abdomen and the lesions we are trying to detect even smaller. This is the challenge of equine MRI.
The relative amount of signal generated from any volume of tissue is very important and is known as the signal to noise ratio or SNR. Basically, low field magnets cause less protons to realign with the main magnet field and this results is less signal being generated from a given volume of tissue. Low SNR results in an inferior image. This is compensated for by making the slices thicker and repeating the pulsing sequence a number of times (known as increasing the number of averages, excitations or acquisitions). Thicker slices potentially results in missing lesions (so called volume averaging artifact). Increasing scanning time results in a higher change of the horse moving and ruining the study if in a standing system, or more anesthesia time in a low field recumbent system.
There is absolutely no doubt, that the best way to optimize an MRI study of a horse’s limb, is to acquire high contrast, high resolution, thin slice images in as short a time as possible. This is best achieved by using a high field and eliminating voluntary motion by general anesthesia.