Diffusion Tensor Distribution ( DTD ) MRI

Measuring and mapping the diffusion tensor distribution (DTD) via MRI holds the promise of revealing the tissue microstructure at sub-voxel resolution. Recent approaches have used the covariance of diffusion tensor as a means to characterize DTD within a voxel [1], [2]. It can be shown that several terms in the covariance matrix are not observable when using standard rank-1 b-matrix acquisition schemes using single pulsed field gradient (PFG) MR sequences, emphasizing the need for higher-rank b-matrix acquisitions for DTD imaging. Several pulse sequences have been introduced for higher-rank b-matrix encoding [2]–[4] but their applicability is limited by their long echo times and/or the lack of well-defined diffusion times. A new efficient, easy to implement b-matrix encoding strategy is presented which is capable of generating ranks 1, 2 and 3 b-matrices with well-defined diffusion times. The presented method is also immune to concomitant field errors which are thought to confound DTD estimation [5]. The developed method is tested on a macroscopic and microscopically isotropic polydimethylsiloxane (PDMS) phantom, and on excised rat brain tissue.