Abstract #0285

Kholmovski E, Parker D

GRAPPA, widely used technique for parallel MRI, has serious limitations for high reduction factors. Recently, new algorithms have been proposed to resolve this GRAPPA limitation. These algorithms exploit all available data dimensions to find reconstruction coefficients allowing more reliable reconstruction for high reduction factors. However, inclusion of additional data dimensions in the reconstruction substantially increases the number of reconstruction coefficients resulting in low computational efficiency of these new algorithms. A novel algorithm has been developed that achieves equivalent image quality for high reduction factors as the multi-dimensional auto-calibrating techniques but has computational efficiency comparable with the computational efficiency of GRAPPA.