Analysis of new diffusion tensor imaging anisotropy measures in the three-phase plot

Abstract

Purpose: To evaluate diffusion anisotropy from diffusion tensor imaging using new measures derived from Hellinger divergences and from compositional data distances. Materials and Methods: New anisotropy measures obtained from the diffusion tensor imaging were measured and compared with classic ones such as fractional anisotropy (FA) and relative anisotropy (RA). The evaluation was done using the three-phase plot (3P-plot). The measures were compared with regard to their sensitivity to detect white and gray matter changes on human DTI data acquired from five normal volunteers. For each volunteer, different volumes of interest located in white matter (WM) and gray matter (GM) were considered. Results: The proposed Compositional Kullback-Leibler (KLA) and the classic FA had a similar behavior, although KLA detected better the transitions between white and gray matter. Moreover, KLA showed a better discrimination in areas with great confluence of fibers. Conclusion: KLA detects better than FA the difference between WM and GM. This leads KLA to be a good measure for segmenting WM from GM. © 2010 Wiley-Liss, Inc.