Premium
TU‐G‐134‐06: Quantitative DCEMRI of the Breast at 1.5T and 3T
Author(s) -
Pineda F,
Medved M,
Fan X,
Ivancevic M,
Newstead G,
Abe H,
Sennett C,
Karczmar G
Publication year - 2013
Publication title -
medical physics
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 1.473
H-Index - 180
eISSN - 2473-4209
pISSN - 0094-2405
DOI - 10.1118/1.4815484
Subject(s) - gadodiamide , calibration , nuclear medicine , signal (programming language) , biomedical engineering , materials science , breast mri , dynamic contrast , medicine , calibration curve , nuclear magnetic resonance , magnetic resonance imaging , mathematics , breast cancer , detection limit , physics , statistics , computer science , radiology , mammography , cancer , programming language
Purpose: To investigate whether quantitative methods reduce the variability in dynamic contrast‐enhanced MRI (DCEMRI) of the breast in patients scanned at two field strengths. Methods: We designed calibration phantoms, consisting of compartments with varying concentrations of gadodiamide solutions (0.05‐0.5mM Omniscan), that were placed into a 16‐channel bilateral breast coil during the DCEMRI. Eleven patients were scanned at both 1.5T and 3T (Philips Achieva 1.5T/3T‐TX) under an IRB approved protocol. The phantoms provide reference signals used to calibrate the signal vs. concentration relationship, leading to a time‐series displaying the concentration of contrast media. Signal enhancement and concentration curves were fit to an empirical mathematical model (EMM) which has been shown to accurately describe kinetic curves in tissue. The parameters from the EMM were compared for the signal enhancement and concentration in enhancing lesions. Results: For most of the parameters in the EMM conversion to concentration reduced the mean difference between 1.5T and 3T, however only uptake slope saw a significant reduction (54% vs. 20%, p=0.02). Overall washout rate saw an increase in variation after conversion. This could be due to increased error in the estimation of concentration at higher values, an issue currently being addressed. Parameters descriptive of contrast media uptake were more reproducible. Plots of signal enhancement versus concentration showed a single value of enhancement corresponding to a range of concentration, suggesting enhancement alone does not provide an accurate measure of concentration. Conclusions: Quantitative methods reduced the variability of dynamic parameters across two fields. Refinement of these methods may further reduce this variability, including a more accurate measurement of proton density ratio between tissue and the phantoms. Parameters descriptive of uptake were more reproducible, suggesting uptake should be a primary diagnostic variable. Quantitative methods have the potential of providing standardized information, which could be used in longitudinal or inter‐institutional studies. Funding: Philips Healthcare. Department of Defense Breast Cancer Research Program Predoctoral Traineeship Award. Illinois Department of Public Health