Simultaneous measurement of T /B and pharmacokinetic model parameters using active contrast encoding (ACE)-MRI.

TitleSimultaneous measurement of T /B and pharmacokinetic model parameters using active contrast encoding (ACE)-MRI.
Publication TypeJournal Article
Year of Publication2017
AuthorsZhang J, Winters K, Reynaud O, Kim SGene
JournalNMR Biomed
Volume30
Issue9
Date Published2017 Sep
ISSN1099-1492
KeywordsAnimals, Cell Line, Tumor, Computer Simulation, Contrast Media, Feasibility Studies, Magnetic Resonance Imaging, Mice, Inbred BALB C, Mice, Inbred C57BL, Models, Theoretical, Numerical Analysis, Computer-Assisted, Phantoms, Imaging, Uncertainty
Abstract

The aim of this study was to assess the feasibility of combining dynamic contrast enhanced-magnetic resonance imaging (DCE-MRI) with the measurement of the radiofrequency (RF) transmit field B and pre-contrast longitudinal relaxation time T . A novel approach has been proposed to simultaneously estimate B and T from a modified DCE-MRI scan that actively encodes the washout phase of the curve with different amounts of T and B weighting using multiple flip angles and repetition times, hence referred to as active contrast encoding (ACE)-MRI. ACE-MRI aims to simultaneously measure B and T , together with contrast kinetic parameters, such as the transfer constant K , interstitial space volume fraction v and vascular space volume fraction v . The proposed method was tested using numerical simulations and in vivo studies with mouse models of breast cancer implanted in the flank and mammary fat pad, and glioma in the brain. In the numerical simulation study with a signal-to-noise ratio of 10, both B and T were estimated accurately with errors of 5.1 ± 3.5% and 12.3 ± 8.8% and coefficients of variation (CV) of 14.9 ± 8.6% and 15.0 ± 5.0%, respectively. Using the same ACE-MRI data, the kinetic parameters K , v and v were also estimated with errors of 14.2 ± 8.3% (CV = 13.5 ± 4.6%), 14.7 ± 9.9% (CV = 13.3 ± 4.5%) and 14.0 ± 9.3% (CV = 14.0 ± 4.5%), respectively. For the in vivo tumor data from 11 mice, voxel-wise comparisons between ACE-MRI and DCE-MRI methods showed that the mean differences for the five parameters were as follows: ΔK  = 0.006 (/min), Δv  = 0.016, Δv  = 0.000, ΔB  = -0.014 and ΔT  = -0.085 (s), which suggests a good agreement between the two methods. When compared with separately measured B and T , and DCE-MRI estimated kinetic parameters as a reference, the mean relative errors of ACE-MRI estimation were B  = -0.3%, T  = -8.5%, K  = 11.4%, v  = 14.5% and v  = 4.5%. This proof-of-concept study demonstrates that the proposed ACE-MRI method can be used to estimate B and T , together with contrast kinetic model parameters.

DOI10.1002/nbm.3737
Alternate JournalNMR Biomed
PubMed ID28544159
PubMed Central IDPMC5557664
Grant ListP30 CA016087 / CA / NCI NIH HHS / United States
P41 EB017183 / EB / NIBIB NIH HHS / United States
R01 CA160620 / CA / NCI NIH HHS / United States
Related Institute: 
MRI Research Institute (MRIRI)

Weill Cornell Medicine
Department of Radiology
525 East 68th Street New York, NY 10065