Meeting Banner
Abstract #0530

Spectroscopy with Linear Algebraic Modeling (SLAM): Speed and Quantification in Brain Tumor Studies

Yi Zhang1, 2, Refaat E. Gabr1, Jinyuan Zhou1, 3, Robert G. Weiss1, 4, Paul A. Bottomley1, 2

1Division o MR Research, Department of Radiolgoy, Johns Hopkins University, Baltimore, MD, United States; 2Electrical and Computer Engineering, Johns Hopkins University, Baltimore, MD, United States; 3F. M. Kirby Research Center for Functional Brain Imaging, Kennedy Krieger Institute, Baltimore, MD, United States; 4Division of Cardiology, Department of Medicine, Johns Hopkins University, Baltimore, MD, United States

The inclusion of potentially useful MRS acquisitions in clinical MRI exams is often precluded by long study times for chemical shift imaging (CSI). Global- or lesion-averaged MRS-measurements can usually suffice for assessing metabolic status. A recently proposed methodspectroscopy with linear algebraic modeling (SLAM)could provide such assessments, in addition to a many-fold speed-up in scan-time. Here, SLAM applied retroactively to patients with brain tumors, is shown to yield quantitatively indistinguishable results from conventional 2D 1H CSI with an acceleration factor of six. Proactive studies demonstrate comparable results to CSI with a speedup factor of 14.

Abstract PDF Presentation Video

Keywords

acquisition acquisitions agree agreement algebraic almost applied apply assigned audience background blue brain brown cardiology central chemical choline clinical clinicians compartment compartmental compartments computer cost defined desired division electrical enable encoded encodes engineering excluding extended faster five fold fraction functional grant grid heart identical implementation implemented incorporating institute johns lesion lesions linear little long many medicine modeling much newly noise often original otherwise overlaid pathologies patients potentially precluded prior proactive proposed protocol quantification quantified quantitative reconstruct reconstructed reconstruction register retroactive retroactively scalp scanner scientists scout segment segmentation segmented slam space spectra spectroscopy spectrum speed speedup step studies successfully support target tumor tumors undergoing useful user vertical volunteer volunteers whole yielded