Abstract
Interpolation is a necessary step in magnetic resonance (MR) angiography when the center-to-center spacing between the MR slices is larger than the in-plane pixel size of the slices. Conventional methods interpolate normal to the slice only, causing a staircase artifact for oblique vessels. Reconstructions of these vessels demonstrate jagged structures with reduced resolution and contrast. A directional interpolation scheme is presented here, in which the orientation of interpolation is determined locally by template matching. By aligning this orientation with the vessels, the algorithm reconstructs the correct shape and contrast for oblique vessels. Simulations and application of the algorithm to clinical MR angiography data demonstrate the advantages of directional interpolation over conventional methods. Examples are shown of combining the algorithm with background suppression, where the background values in the slices are suppressed prior to projection and directional interpolation. This results in even higher vessel-to-background contrast.

This publication has 13 references indexed in Scilit: