Image contrast using the secondary and tertiary eigenvectors in diffusion tensor imaging.

Abstract

Diffusion tensor imaging (DTI) is a new imaging modality that can provide unique information on brain white matter anatomy. Measurements of water diffusion constant along multiple axes are fitted to a tensor model, from which the diffusion anisotropy and dominant fiber orientation can be estimated. Even though the tensor model is an oversimplification of the underlying neuroanatomy, information within the tensor has not been fully utilized in routine research and clinical studies. In this study we proposed and examined the properties and anatomical contents of several DTI-derived image contrasts that utilize all three eigenvectors. The new contrasts are studied and validated using known anatomical structures in ex vivo mouse brain and embryonic mouse cortex. Application to human white matter is illustrated. Our results suggest that when these contrasts are combined with a priori anatomical knowledge, they reveal neuroanatomical information that is useful for tissue segmentation and diagnosis of white matter lesions.