Method to significantly reduce bias and variance of diffusion anisotrophy measurements

A method for assessing diffusion anisotropy in an object that was subjected to resonance imaging, said method comprising the computer executed steps of:

• obtaining information signals resulting from the resonance imaging, said information signals sufficient for representing at least one eigenvalue of a diffusion tensor for each of a plurality of localized regions in said object, each of the at least one eigenvalue representing diffusively of spin-labelled species displacement in the object along a corresponding rotationally invariant principal direction of spin-labeled species displacement;
• and
• generating for at least one localized region an improved estimate of a function of the at least one eigenvalue of the diffusion tensor, based on said information signals and on a representation of local structural direction field coherence or order in the object.

A computer program embodied on a computer-readable medium, the computer program used for assessing diffusion anisotropy for displacements of spin-labeled species in an object that was subject to resonance imaging, the computer program comprising:

• a code segment to obtain information signals resulting from the resonance imaging, the information signals sufficient for representing at least one eigenvalue of a diffusion tensor for each of a plurality of localized regions in said object, each of the at least one eigenvalue representing diffusively of spin-labelled species displacement in the object along a corresponding rotationally invariant principal direction of spin-labeled species displacement;
• and
• a code segment that generates for at least one localized region an improved estimate of a function of the at least one eigenvalue of the diffusion tensor, based on said information signals and on a representation of local structural direction field coherence or order in the object.