Speaker: Ed DiBella

Title: Compressed sensing (CS) refers to methods for reconstructing data that are highly compressible, using random sampling and L1 norms.

Following on the heels of Fred's excellent JC on CS, I stumbled on a paper that a group at Stanford has submitted using CS for MRI.

Reference: "Sparse MRI: The application of compressed sensing for rapid MR imaging" Lustig, Donoho, and Pauly, submitted to Magn Reson Med

The paper is at a central CS website. (The website is well organized with links to many CS papers and to software.) Or directly.

The approach is in many ways similar to the constrained reconstruction approach that we are currently developing. Our approach stems from initial work done with Oleg Portniaguine when he was at SCI ("Inverse methods for reduced k-space acquisition," Portniaguine, Bonifasi, DiBella, and Whitaker, presented at /ISMRM/, pp. 481, Toronto, Canada, 2003). The constrained reconstruction approach is also the topic of our current grant proposals, and was recently accepted to Magn Reson Med ("Temporally Constrained Reconstruction (TCR) of Dynamic Cardiac Perfusion MRI". Adluru, Awate, Tasdizen, Whitaker, and DiBella). Interestingly enough, I'm also reviewing a paper for PMB that uses CS for MRI. I plan to present the sparse MRI approach and compare and contrast it to our methods for MRI reconstruction.