Abstract
This study describes a method of utilizing unaveraged dipolar effects to characterize and compute collagen fiber tracks using magnetic resonance imaging. The technique yields information about fiber structure with some similarities to what can be obtained in brain using diffusion tensor imaging, but relies on a completely different physical mechanism, namely, unaveraged homonuclear dipolar interactions. The method is probably only appropriate for highly ordered collagen rich tissues. A goat knee meniscus was embedded in a spherical epoxy ball and the magnetic resonance signal intensity was examined as a function of sample orientation to a 3T static field using a three‐dimensional gradient echo sequence. Unaveraged dipolar interactions caused a sixfold signal variation with orientation. After correction for coil sensitivity and registration of the images, a principal dipolar direction was computed for each voxel. The data were analyzed and viewed as dipolar direction and standard deviation (anisotropy) maps. Circumferential fibers within the meniscus were visualized as fiber tracks grown using diffusion tensor imaging software. The acronym dipolar anisotropy fiber imaging is proposed for this technique. Magn Reson Med, 2011. © 2010 Wiley‐Liss, Inc.