A case study of hyperthermia induced temperature computations in human sarcomas using discrete vasculature and relative perfusion maps

One of the most important technical aspects in clinical hyperthermia is the ability to measure and/or simulate the 3D temperature fields. Related to that, an essential part is the way in which the complex heat transfer related to vasculature is described. We report here the results of a collaboration between the hyperthermia modeling groups from the UMC Utrecht, The Netherlands, and Duke UMC, USA. Utrecht's hyperthermia group has developed a flexible, discrete vasculature thermal model (DIVA) (Kotte et al. 1996) that describes the heat transfer related to discrete vasculature. The vasculature was imaged using MR angiography. To account for the smaller vessels that are responsible for the significant bioheat transport, relative perfusion maps measured at Duke using dynamic enhanced-magnetic resonance imaging were used. Alternatively, the VAMP program (Van Leeuwen et al. 1998) was used to artificially generate smaller vasculature. The cases with discretized vasculature were compared to continuum models where either heterogeneous isotropic perfusion, or relative perfusion maps were used. All simulations were compared to MR thermometry data. The conclusion is that for tumors crossed by or near large vessels, a combination of large vessels discretization and perfusion maps yields temperatures that match very well the MR thermometry data.

Duke Scholars

Author Shiva Kumar Das Radiation Oncology