Anatomical validation of diffusion tensor imaging (DTI)

Abstract

Listen

Translate article

The Magnetic Resonance Image (MRI) based Diffusion Tensor Imaging (DTI) technique reveals brain white matter architecture in vivo by measuring the direction of the diffusion of water molecules along the anisotropic fiber bundles. Validation of DTI at present is limited to qualitative comparisons of previously known connections in the common normal human brain. Anatomical validation of DTI through histological analysis ensures that each image is representative of the inherent white matter architecture, and improves algorithms that create a tractographic image from the tensor signals in each voxel of the image. Human brain samples were imaged at 3 Tesla (T) (2.5 mm voxels) followed by sectioning and histologic processing of the tissue. Results comparing the histologic data to the imaging show good correspondence in white matter areas between both fiber integrity and axonal orientation. Vector coordinates drawn from the tissue image match the principal fiber direction derived from the probability density function of the corresponding image voxel. We anticipate that this will provide a valuable resource for the validation of DTI methods.Grant Funding Source: NIH/NIA 1 R21 AG037843‐02, NIH/NCRR P41 RR013642‐12S1, Translational Research Fund (UCLA), and the McGinty Family Foundation