WITHDRAWN: Improving Clinical Utility of Physiologic MRI

• Author(s): John Kornak and Karl Young and Ying Lu and Norbert Schuff and Jeff Kasten and Mike Weiner
• Keywords: MRI ; K-Bayes ; Bayesian Image Analysis ; k-space ; reconstruction ; DFT

Abstract

A statistical procedure (K-Bayes) is presented for the improved reconstruction of physiological Magnetic Resonance Imaging (MRI) of the brain: specifically, perfusion MRI for measuring blood perfusion levels and magnetic resonance spectroscopic imaging for quantifying multiple metabolite levels. K-Bayes improves image quality and spatial resolution while overcoming inherent limitations of conventional discrete Fourier transform (DFT) based reconstruction: Gibbs ringing, partial voluming and aliasing. K-Bayes achieves these improvements by 1) using high-resolution anatomical prior information from structural MRI to provide constraints for the physiological MR process; and 2) simultaneously relating the physiological image to be reconstructed to the sampled data points in k-space (frequency space). K-Bayes reconstruction leads to more accurate and precise measurements in individual brain r