Reconstructing an Increased Number of Simultaneously Excited fMRI Slices

• Activity Number: 347
• Type: Contributed
• Date/Time: Tuesday, August 5, 2014 : 10:30 AM to 12:20 PM
• Sponsor: Section on Statistics in Imaging
• Author(s): Iain Bruce*+ and Daniel Rowe
• Companies: Marquette University and Marquette University
• Keywords: functional connectivity MRI ; image reconstruction ; neuroimaging ; parallel MRI ; induced correlation

Abstract

In parallel MRI, multiple coils are placed around an object to concurrently sample the spatial frequency spectrum of the object. With an overlap of coil magnetic fields, there is an excess of spatial information in each pixel of the object image. This allows for data acquisition to be accelerated by simultaneously acquiring spatial frequencies for multiple slices of the volume being imaged with each coil. By separating the aliased slices through a least squares estimation, the acceleration factor in parallel MRI is typically limited to the number of coils. This study expands upon a means of incorporating additionally acquired calibration data together with coefficients of orthogonal polynomials to improve the rank of the design matrix used for slice separation. It is shown that accelerations in excess of the number of coils can be achieved, and the statistical implications of acquiring and separating multiple slices at once are described.