--traditional MRI, even with the different type of images, is not that good at picking up lesions in the dorsal root ganglia; the radiologist reading the scans has to additionally be looking for them very specifically and to be familiar through training what they might look like. Otherwise they just seem to be signal fluctuations of uncertain significance.
What I believe they may actually have been using in this study, although I don't believe it was referred to this way, then, is the fairly new specialized MRI technique known as magnetic resonance neurography or MRN. This is a farily new adjustment, not done at all centers, of the MRI protocol using magnetic wave passes designed to look at nerves more closely and selectively,and this seems to create more detailed images of nerve structures than was previously possible.
We've had a few discussions about MRN here that you can search up--this article, though technical, gives a pretty good idea what it involves:
http://www.ncbi.nlm.nih.gov/pmc/articles/PMC3622412/