This invention describes the apparatus and data processing technique for correcting the noise based systematic bias within MRI diffusion tensor image (DTI) sequences.
About
Description This invention describes the apparatus and data processing technique for correcting the noise based systematic bias within MRI diffusion tensor image (DTI) sequences. The specific claim requested in this IP document is for coverage of 1) the phantom design, 2) the concept of the parametric remapping technique, and 3) the use of the phantom to determine the parametric map. During a DTI sequence, a series of images is generated by applying gradient magnetic fields along particular directions, to measure the amount of diffusion of water within tissues in those same directions (or whatever may be in the scanner, such as a reference test object known as a "phantom"). These sequences are useful for characterizing tissue substructure (for example, cellular size, membrane integrity, etc), which may change with the introduction and progression of disease. The technique of DTI characterizes the directional dependence of diffusion using a tensor, which specifies the amount of diffusion (for example, eigenvalues) along three principle axes (for example, eigenvectors) of an ellipsoid. Other clinical metrics, such as fractional anisotropy (FA), are derived from these two sets of quantities, generally describing how directionally uniform diffusion proceeds in tissues and along what directions the non-uniform diffusion may occur.