Joshua P. Yung1, 2, David Fuentes1, John D. Hazle1, 2, Jeffrey S. Weinberg3, R. Jason Stafford1, 2
1Department of Imaging Physics, The University of Texas M.D. Anderson Cancer Center, Houston, TX, United States; 2The University of Texas Graduate School of Biomedical Sciences, Houston, TX, United States; 3Department of Neurosurgery, The University of Texas M.D. Anderson Cancer Center, Houston, TX, United States
Synopsis. Minimally-invasive MR-guided thermal therapies have been combined with model-based filters to improve precision and robustness of MR thermometry. When using a model, such as the bioheat transfer equation, to provide temperature estimate predictions, the propagation of the covariance matrix becomes computationally intensive. In this work, a characterization study was performed to investigate the effect of localization and model error covariance in order to provide a reduction in computational complexity while in the presence of simulated artifacts. Dice similarity coefficient and RMS error was used to evaluate the temperature model-based Kalman filter temperature estimates with MRTI and post-treatment imaging.