Meeting Banner
Abstract #0077

Blind Compressed Sensing Dynamic MRI

MAGNA25Sajan Goud Lingala1, Mathews Jacob2

1Biomedical Engineering, The University of Iowa, Iowa city, IA, United States; 2Electrical and Computer Engineering, The University of Iowa, IA, United States

In this work, we introduce a novel blind compressive sensing frame work for dynamic MRI reconstruction. This models the temporal profile at each voxel as a sparse linear combination of temporal basis functions chosen from a large dictionary, which are also estimated from the data. We show that the model significantly reduces the number of degrees of freedom than what is seen in schemes based on promoting low rank structure of the data. We demonstrate this concept on myocardial perfusion data sets with significant inter-frame motion. Significant improvement in the reconstruction qualities over low rank schemes are observed (eg: better preservation of subtle spatial details, reduced temporal blur and artifacts).

Keywords

acceleration accelerations accurately achieved acquisitions approximate approximation arrows artifacts basis biomedical blind blurring borders breathing challenging coefficient coefficients columns combination compressed conjugate consider constraints correspond corresponds dataset datasets degrades depicts details dictionary dotted dwarf dynamic energy enforce engineering equivalently error estimation example exploit factorization features fine flash fold form frame frames free fully functions funding gating good gradually greedy heavy huge inter introduce introduced issues iterates lack larger linear local loss makes matrix minimal minimization minimize model motion much muscles myocardial nature norm note novel nuclear observe often papillary parameterized pattern penalty perfusion picked pose problem problems profile profiles promoting quadratic quality randomly rank reconstruction reconstructions reduced regularizations reported represent representation respectively respiratory retrospectively review rows sample sampled sampling saving scanner scheme schemes sensing sensitive significantly similarity since slow solution solvers solving space spaced sparse sparsity spatial specifically specified start steps strategy temporal termed trajectory trans trends uniformly ventricles white zero