Meeting Banner
Abstract #0241

Comparison of Signal Intensity & Standard Techniques for Estimation of Pharmacokinetic Parameters in DCE-T1 Studies of Glioblastoma: Using Model Selection

Hassan Bagher-Ebadian1, 2, Siamak P. Nejad-Davarani1, 3, Rajan Jain4, Douglas Noll3, Quan Jiang1, Ali Syed Arbab4, Tom Mikkelsen5, James R. Ewing1, 2

1Neurology, Henry Ford Hospital, Detroit, MI, United States; 2Physics, Oakland University, Rochester, MI, United States; 3Biomedical Engineering, University of Michigan, Ann Arbor, MI, United States; 4Radiology, Henry Ford Hospital, Detroit, MI, United States; 5Neurosurgery, Henry Ford Hospital, Detroit, MI, United States

DCE-pharmacokinetic models rely on the construction of an observation equation which demands conversion of the measured signal intensity (SI) profile into an indicator concentration time-course. Recent studies have proposed that the normalized SI [(St-S0)/S0] be used instead of the longitudinal-relaxation-rate change (ΔR1) in DCE-MRI permeability analyses. However, we know of no assessment of the agreement in the estimated permeability parameters using different measures of concentrations (SI-ΔR1). The goal of this study is to evaluate the use of SI, as opposed to ΔR1, in the estimation of permeability parameters in DCE-T1-3D-Spoiled-Gradient-Echo studies in the brains of ten treatment-nave patients with glioblastoma.

Abstract PDF Presentation Video

Keywords

accordingly administration agreement analyses analytical analytically analyzed appropriate around array arterial assessment assumed begun biased biasing biomedical brains channel characterize chosen coil concentration conducted confirms constant construct constructed contrast conversion core course criteria curve curves demands describe dose dynamic eight enclosed engineering enhanced enhancing equation error establish estimating estimation excellent excite exemplary experimental expression extracellular final fits fitted fitting flow ford forward function goal gradient gray henry hospital illustrates imply indicator injected injection input instead intensity interstitial invest know leakage least linear longitudinal manually many mapping maps math matrix measured measures model models nested neurology neurosurgery normalized opposed optimization pathology patient patients perfusion permeability physics physiological plasma population post prior procedure process processing profile properties proposed radiology receiver recent reduced regional relate respectively reverse selection served sets slices space squared squares starting statistical studies system theory tissue trace trans transfer treatment turn typical underestimated utilizes variable variation vascular volume volumes white