Meeting Banner
Abstract #0236

Comparison of Arterial Input Functions by Magnitude & Phase Signal Measurement in Dynamic Contrast Enhancement MRI Using a Dynamic Flow Phantom

SUMMA25Sangjune Laurence Lee1, Warren Foltz1, Brandon Driscoll1, Ali Fatemi1, Cynthia Menard1, Catherine Coolens1, Caroline Chung1

1Department of Radiation Oncology, University of Toronto, Toronto, Ontario, Canada

Determining the arterial input function (AIF) is critical for the accuracy of kinetic modeling of DCE MRI measures. However, the magnitude signal derived AIF suffers from in-flow, slice profile, and T2* effects. Phase signal has been shown to be an alternative method of acquiring the AIF data. Here, we evaluate the accuracy of AIFs derived from the magnitude and phase signal in a dynamic flow phantom, providing a controlled, gold-standard framework. The phase-derived AIF approached actual peak Gd-DTPA concentrations within all imaging slices and at tested flow rates up to 7.5 mL/s, while the magnitude-derived AIF was grossly attenuated.

Abstract PDF Presentation Video

Keywords

accuracy accurate achieved acquisitions affected allow analyzed approximately array arterial artifact artifacts attenuated attenuation better blood blunt bolus calibration centrifuge characterization clinical closest coil compatibility compensation concentration concentrations conditions connected constant contrast controlled correct corrected critical cross dependence dependent derived developed dilution dilutions direct displacement dominated dynamic edge edges elevation enhancement environment estimation expected experiment flash flow fluctuating fluid fold function functions glycerol graphs house identify injected injection injector input inside intensity introduce investigate investigation limiting linear magnitude manner matched measured measures middle mimic mixture models modifications parallel passage peak peaks performances permeability phantom physiological pixel placed plateau polarization position positive power predict prescribed previous probably profile profiles programmed pump quantification quantify radiation reductions related reliable repeated respectively robust robustness section severe slice slices spatial spine stable stack static susceptibility system systematic technologies temporal towards transient tube tubing tumor typically upstream utilized validate validated variable variation varying vascular velocities volume water