Kaung-Ti Yung1, Shang-Yueh Tsai2, Fa-Hsuan Lin3,4, Pierre-Gilles Henry5, Akio Yoshimoto6, Stefan Posse7,8
1Neurology, University of New Mexico, Albuquerque, NM , USA; 2Electrical and Computer Engineering, National Taiwan University, Taipei, Taiwan, Taiwan; 3Athinoula A. Martinos Center for Biomedical Imaging, MGH-HMS-MIT, Charlestown, MA, USA; 4Radiology, Massachusetts General Hospital, Boston, MA, USA; 5Center for Magnetic Resonance Research and Department of Radiology, University of Minnesota, Minneapolis, MN, USA; 6Electrical and Computer Engineering, University of New Mexico, Albuquerque, NM, USA; 7Neurology, University of New Mexico, Albuquerque, NM, USA; 8Electrical and Computer Engineering, University of New Mexico, Albuquerque, NM , USA
In this study we compare TE-averaged with short TE (15 ms) PEPSI data acquired in human brain at 3T, using identical voxel size (1 cc). Short TE data have higher SNR and Cramer-Rao lower bounds for Glu are lower compared to the TE-averaged data. The number of voxels above threshold for TE-averaged data is smaller than for short TE data. Short TE acquisition is advantageous for clinical studies for sensitivity reasons and due to the shorter measurement time. TE-averaged acquisition is complementary to short TE acquisition for identification of Glu albeit at lower spatial resolution.