Shawn Joseph Kisch1, Philip M. Robson2, Catherine DG. Hines3, Bryan T. Addeman1, Huanzhou Yu4, Ann Shimakawa4, Ananth Madhuranthakam4, Ajit Shankaranarayanan4, Jean Brittain4, Scott Reeder3, Charles McKenzie1,5
1Department of Medical Biophysics, University of Western Ontario, London, ON, Canada; 2Department of Radiology, Beth Israel Deaconess Medical Center and Harvard Medical School, Boston, MA, USA; 3Department of Biomedical Engineering, University of Wisconsin-Madison, Madison, WI, USA; 4Applied Sciences Laboratory, GE Healthcare; 5Department of Biomedical Engineering, University of Western Ontario, London, ON, Canada
We investigate the variation in signal-to-noise-ratio (SNR) in parallel MRI accelerated IDEAL images due to the k-space sampling pattern. The pseudo replica SNR measurement used allows for an accurate measurement of SNR in both liver water and liver fat images produced by the IDEAL sequence. A comparison between different sampling patterns in each IDEAL echo with an overall acceleration factor of two in a PMRI sequence is made. The results show that the SNR is independent of the choice of calibration echo, but is maximized when a uniform k-space sampling pattern is used.