Rapid cardiac MRI using random radial trajectories

We introduce using sparse MRI with random radial trajectories in order to allow MRI images reconstruction from a small number of acquired k-space data, so fulfilling the demands of dynamic MRI. Different random radial trajectories are generated by varying the parameters of the added random perturbations to the radial trajectories. Images reconstruction is performed using the non-linear L1 norm reconstruction. Entropy is computed for the reconstructed images, as a quantitative measure for the reconstructed image quality. Both phantom simulation and real cardiac images are used in our experiments. Our results show that more sparsely sampled images can be reconstructed with higher quality compared to using non-randomly sampled radial kappa-space trajectories.