Meeting Banner
Abstract #0283

Initial in vivo Validation of a Hybrid Magnitude/Complex MRI-Based Method for Liver Fat Quantification

Ghaneh Fananapazir1, Xiaodong Zhong2, Lauren M.B. Burke1, Brian M. Dale3, Berthold Kiefer4, Daniel T. Boll1, Mustafa R. Bashir1

1Radiology, Duke University Medical Center, Durham, NC, United States; 2MR R&D Collaborations, Siemens Healthcare, Atlanta, GA, United States; 3MR R&D Collaborations, Siemens Healthcare, Morrisville, NC, United States; 4MR R&D Collaborations, Siemens Healthcare, Erlangen, Germany

Hepatic steatosis can precede liver inflammation, chronic fibrosis, and ultimately liver failure and malignancy. A novel method has been developed to measure proton density fat fraction from a multi-echo MRI acquisition. This is a hybrid magnitude/complex data method which takes advantage of the insensitivity to phase error of magnitude-data techniques and the full 0-100% dynamic range of complex-data techniques. This study looked to validate this technique against co-localized T2-corrected single voxel spectroscopy. Agreement between these two methods was very strong. Additionally, this technique allows for assessment of the variability of disease within the liver as well as total lipid burden.

Abstract PDF Presentation Video

Keywords

abdominal abstract accuracy acquisition addition additional advantage advantages agreed agreement allowing appears approved arrive assessing assessment audience avoiding board boll briefly burden chosen chronic clinical collaborations collected compartments complex consecutive considered content corrected correction correlation coverage cubic dale density deposition derived derives details determine developed disease distribution duke dynamic echoes emerging error examinations examine example excellent failure fibrosis fitting five flexible fraction fractions full guess guesses hepatic holds hybrid important in vivo includes inflammation inherent initial insensitivity institutional intercept iron leverages linear lipid liver local localize localized magnitude malignancy measured measuring medical methodologies modeling patient patients peak phenomenon placed precede promise prospective proton prove quantification quantifying radiologists radiology rank reached recently reconstruction regression relationship reported representative represents response review seed separate severity signed significance slope solution solutions spectroscopic spectroscopy speed stable staging statistical statistically steps strong system takes target though treatment ultimately update useful utility validate validated validation variability vessels water whether whole