Magnetic resonance imaging (MRI) and nuclear magnetic resonance (NMR) systems have much in common with polarized ion sources.  All of these systems work with the intrinsic atomic and nuclear spins of various elements.  Polarized ion sources concentrate on producing high polarization levels (e.g. alignment of the nuclear spins) in ion beams, NMR systems infer chemical properties from analysis of  the nuclear spins in a sample, and MRI scanners create images from the few parts-per-million polarization that occurs naturally when a collection of protons  is contained in a magnetic field.  

Unfortunately the techniques used to polarize protons outside the human body are generally incompatible with living tissue.  However we note that even a small degree of induced polarization (~0.1%) would enhance the MRI signal by a factor of 1,000!  Such an increase in signal would significantly improve the quality of the images, significantly reduce the time required to acquire an image, or enable magnetic resonance microscopy.  The current approach to improving the signal strength is to increase the stength of the magnetic field.  This approach requires ever more costly magnets and higher operating frequencies whose power is more strongly absorbed by the tissues being imaged.

One approach to polarizing atoms in the body has been developed where highly polarized helium or xenon gas is inhaled by a patient. This technique allows imaging of the lungs, which is not possible with conventional MRI because of the low density (diaphanous nature) of lung tissue and the paramagnetism of the oxygen molecules in the air. The high polarization of the inhaled helium or xenon more than offsets the low gas density and, even though the polarization decays rapidly inside the body, it remains large enough for long enough to produce credible images. The highly polarized gas is produced by optical pumping techniques that have also been used to polarize protons in polarized ion sources.

Tomographic techniques are used to reconstruction of x-ray images in computer-aided tomography (CAT) imaging systems. These techniques can also be applied to other types of diagnostics where the results of a series of one-dimensional measurements can be correlated to produce a 2-dimensional (or even 3-dimensional) functional representation of a particular parameter. For example, an electrical conductivity map of a human heart could be reconstructed from a series of electrodes measuring the propagation velocity of the heart contraction wavefront. This electrical conductivity map could be used to identify trouble spots, scarring, and circular conductivity paths that could lead to future heart attacks. Cardiac ablation is a procedure used to repair such trouble spots, but an accurate map of the trouble area needs to be produced to ensure ablation of the problem area and not of healthy tissue.