Cory Robert Wyatt1, Brian J. Soher2,
James R. MacFall2
1Department of Biomedical Engineering,
Duke University, Durham, NC, United States; 2Department of
Radiology, Duke University Medical Center, Durham, NC, United States
Multi-echo
fat-water separation techniques, such as IDEAL, have been shown to be
effective in measuring temperature changes in fatty tissue, but often make
assumptions that allow them to linearize the model in order to simplify the
computation of a solution. This can result in the addition of significant
bias to the measurement of the temperature and the B0 field offset, both
important parameters to monitor during therapeutic heat applications (tumor
ablation, hyperthermia). In this work,
the bias of a multi-peak IDEAL algorithm (without T2* decay) and a new
nonlinear fitting algorithm is characterized using