Compressed sensing (CS) has been demonstrated to accelerate MRI acquisitions by reconstructing sparse images of good quality from highly undersampled data. Motion during MR scans can cause inconsistencies in k-space data, resulting in strong motion artifacts in the reconstructed images. For CS to be useful in these applications, motion correction techniques need to be combined with the undersampled reconstruction. Recently, joint motion correction and CS approaches have been proposed to partially correct for effects of motion. However, the main limitation of these approaches is that they can only correct for affine deformations. In this work, we propose a novel motion corrected CS framework for free-breathing dynamic cardiac MRI that incorporates a general motion correction formulation directly into the CS reconstruction. This framework can correct for arbitrary affine or nonrigid motion in the CS reconstructed cardiac images, while simultaneously benefiting from highly accelerated MR acquisition. The application of this approach is demonstrated both in simulations and in vivo data for 2D respiratory self-gated free-breathing cardiac CINE MRI, using a golden angle radial acquisition. Results show that this approach allows for the reconstruction of respiratory motion corrected cardiac CINE images with similar quality to breath-held acquisitions. Magn Reson Med 70:504-516, 2013. V C 2012 Wiley Periodicals, Inc.

This study presents the first histopathological validation of CMR and endocardial voltage mapping to define acute and chronic atrial ablation injury, including SI thresholds that best match histological lesion volumes. An understanding of these thresholds may allow a more informed assessment of the underlying atrial substrate immediately after ablation and before repeat catheter ablation for atrial arrhythmias.

In this work we reduce undersampling artefacts in two-dimensional (2D) golden-angle radial cine cardiac MRI by applying a modified version of the U-net. The network is trained on 2D spatio-temporal slices which are previously extracted from the image sequences. We compare our approach to two 2D and a 3D Deep Learning-based post processing methods, three iterative reconstruction methods and two recently proposed methods for dynamic cardiac MRI based on 2D and 3D cascaded networks. Our method outperforms the 2D spatially trained U-net and the 2D spatio-temporal U-net. Compared to the 3D spatiotemporal U-net, our method delivers comparable results, but requiring shorter training times and less training data. Compared to the Compressed Sensing-based methods kt-FOCUSS and a total variation regularized reconstruction approach, our method improves image quality with respect to all reported metrics. Further, it achieves competitive results when compared to the iterative reconstruction method based on adaptive regularization with Dictionary Learning and total variation and when compared to the methods based on cascaded networks, while only requiring a small fraction of the computational and training time. A persistent homology analysis demonstrates that the data manifold of the spatio-temporal domain has a lower complexity than the one of the spatial domain and therefore, the learning of a projection-like mapping is facilitated. Even when trained on only one single subject without data-augmentation, our approach yields results which are similar to the ones obtained on a large training dataset. This makes the method particularly suitable for training a network on limited training data. Finally, in contrast to the spatial 2D U-net, our proposed method is shown to be naturally robust with respect to image rotation in image space and almost achieves rotation-equivariance where neither dataaugmentation nor a particular network design are required.