Introduction: A recent report of EORTC-STBSG phase III randomized trial (2010) describes that regional hyperthermia can improve chemotherapy results for patients with high-risk soft-tissue sarcoma and Krawczyk et al. (PNAS, 2011) have just discovered that mild high temperature can prohibit homologous recombination - a process to repair double strand breaks. The clinical outcomes and laboratory findings support the combination use of hyperthermia therapy (HT) with chemotherapy or radiotherapy. However, uncontrolled heating can cause severe burns of bones and other critical structures. Thus, it is extremely important to quantify heat transport and thermal dose in the human body for safe and accurate HT. Materials and Methods: Custom phantoms were made of materials with relative permittivity equivalent to soft tissue and shapes similar to the patient body with embedded heat reflectors simulating bones and heat exchangers simulating the blood streams. Eight thermistors were placed at various depths to monitor temperature changes during the heating process. The phantoms were heated by microwave surface heaters with adjustable heat power and water bolus. At therapeutic temperatures, thermographs and 3D surfaces at depths of interest were captured using an infrared camera and a 3D camera. Results and Discussion: Initial results without heat exchangers show that spatial temperature distribution varies greatly with the heat power, heating duration, treatment depth, surface shape, bolus thickness, and presence of heat reflectors. Understanding effects of the controllable variables of the heat power, heating duration, and bolus thickness and the intrinsic variables of the body shape, bone, and blood flow on the temporal and spatial temperature distributions in patients would allow us to optimize the heating process and to calculate the effective thermal doses.