Conventional MRI relies on measuring the amplitude of electromagnetic waves coming off of excited protons as they return to their equilibrium inside the magnetic field. Although the frequency of these signals also contains a good deal of information, it has proven difficult to create algorithms that can make sense of the data.
Researchers from the University of British Columbia led by Alexander Rauscher have been studying MRI scans of brains of multiple sclerosis patients with an eye toward spotting mathematical signatures within frequency data that can identify specific tissue types. In the journal Neurology they are now reporting that they managed to spot lesion formation in the brain using their new analysis technique. Because the new methodology doesn’t require any physical upgrades to MRI scanners, it can be easily integrated into existing systems with a software update.

More information:
Study abstract in Neurology: Magnetic resonance frequency shifts during acute MS lesion formation
UBC: New imaging technique holds promise for speeding MS research