The MIDAS Journal logo

The MIDAS Journal

Home

On What Manifold do Diffusion Tensors Live?

Pasternak, Ofer, Verma, Ragini, Sochen, Nir, Basser, Peter J.
School of Computer Science, Tel Aviv University, Israel, 69978.
Publication cover image

Please use this identifier to cite or link to this publication: http://hdl.handle.net/10380/1501
New: Prefer using the following doi: https://doi.org/10.54294/7kibc5
Published in The MIDAS Journal - MICCAI 2008 Workshop: Manifolds in Medical Imaging: Metrics, Learning and Beyond.
Submitted by Ofer Pasternak on 2008-09-12 11:32:52.

Diffusion tensor imaging has become an important research and clinical tool, owing to its unique ability to infer microstructural properties of living tissue. Increased use has led to a demand for statistical tools to analyze diffusion tensor data and perform, for example, confidence estimates, ROI analysis, and group comparisons. A first step towards developing a statistical framework is establishing the basic notion of distance between tensors. We investigate the properties of two previously proposed metrics that define a Riemannian manifold: the affine-invariant and Euclidean metrics. We find that the Euclidean metric is more appropriate for intra-voxel comparisons, and suggest that a context-dependent metric may be required for inter-voxel comparisons.