In this study, a patch-based super-resolution algorithm is proposed, which uses prior knowledge from high-resolution 1H MRI to guide the reconstruction of hyperpolarized 13C images. The algorithm was validated with simulation and phantom, and the results show enhanced spatial resolution, SNR and contrast as well as comparable quantification accuracy to the upsampled images from nearest-neighbor, bilinear and spline interpolation methods. Finally, the proposed algorithm was applied to metabolic imaging of human brain with hyperpolarized [1-13C]pyruvate injection, which significantly improve the image resolution, SNR and contrast while keeping the quantification accuracy.
This abstract and the presentation materials are available to members only; a login is required.