Techniques for automatic spinal cord histology characterization for validation of diffusion tensor imaging

F Golabchi; W Hoge; H Mamata; S Maier; D Brooks

doi:10.1109/IEMBS.2004.1403496

Techniques for automatic spinal cord histology characterization for validation of diffusion tensor imaging

Conf Proc IEEE Eng Med Biol Soc. 2004:2004:1640-3. doi: 10.1109/IEMBS.2004.1403496.

Authors

F Golabchi¹, W Hoge, H Mamata, S Maier, D Brooks

Affiliation

¹ Department of Electrical and Computer Engineering, Northeastern University, MA, USA.

PMID: 17272016
DOI: 10.1109/IEMBS.2004.1403496

Abstract

One current focus in diffusion tensor imaging is the validation of DT-MR measurements against histological slices. Current methods require manual correlation by visual inspection. We present algorithms which automatically analyze spinal cord histology slices based on Fourier and morphological operators. They accurately discriminate regions dominated by in-plane caudal-cranial fibers from in-plane axial fiber dominated regions, and can then be used to validate angular direction, anisotropy, and relative eigenvalue estimates from DT-MR.