2019/02/01 by Thomas W. Sanchez, Baran Gözcü, Sanchez, Thomas +11 · 1 citation
Engineering · Medicine · #Advanced MRI Techniques and Applications #Computer Vision and Pattern Recognition (cs.CV) #FOS: Computer and information sciences #FOS: Electrical engineering #Image and Video Processing (eess.IV) #Medical Imaging Techniques and Applications #Sparse and Compressive Sensing Techniques #electronic engineering #information engineering
paper · pdf · doi:10.48550/arxiv.1902.00386
openalex publication_date 2019/02/01 · openalex created_date 2022/07/29 · openalex updated_date 2026/07/28
Compressed sensing applied to magnetic resonance imaging (MRI) allows to\nreduce the scanning time by enabling images to be reconstructed from highly\nundersampled data. In this paper, we tackle the problem of designing a sampling\nmask for an arbitrary reconstruction method and a limited acquisition budget.\nNamely, we look for an optimal probability distribution from which a mask with\na fixed cardinality is drawn. We demonstrate that this problem admits a\ncompactly supported solution, which leads to a deterministic optimal sampling\nmask. We then propose a stochastic greedy algorithm that (i) provides an\napproximate solution to this problem, and (ii) resolves the scaling issues of\n[1,2]. We validate its performance on in vivo dynamic MRI with retrospective\nundersampling, showing that our method preserves the performance of [1,2] while\nreducing the computational burden by a factor close to 200.\n