Dose Analysis by Radiation Treatment Planning System (TPS) Software Vs. Thermoluminescent Dosimeters Output

Solid tumors are often treated with external beams of photons generated by bremsstrahlung radiation. These beams are shaped and filtered to optimize dose to specific regions defined by the treatment plans, which may involve irradiations from multiple angles. It is important that doses to healthy tissue not exceed tolerance doses and that the dose to the tumor be maximized. To accomplish these objectives, commercially available three-dimensional treatment planning software (TPS) is used to calculate doses to healthy tissue and to the tumor. It is generally believed that these commercial software packages calculate doses to the patient to within a few percent. The main objective of this study is to examine this claim. In order to determine the validity of the dose, calculations obtained from the TPS are compared with the dose response recorded by thermoluminescent dosimeters (TLD' s) that were initially introduced in an Anthropomorphic Radiation Therapy (ART) phantom. Before inclusion into the ART Phantom, forty-two TLD's were properly calibrated using an NIST Co-60 source. The calibration of the TLD's took place at the United States Army Dosimetry Center, Redstone Arsenal, Alabama.

The radiation therapy system utilized in this study is the linear accelerator (LINAC) located at Baptist Cancer Center, Knoxville, TN. The LINAC was set to establish a prescription dose of 1 80 cGy to the pelvic region of the ART Phantom where the mock tumor was located. Analysis of the data reveals that 37 percent of the data calculated by the TPS differ from the dose response by the TLD-IOOs by more than 5%, with the majority of difference taking place at locations close to the emulated tumor, which seem to be due to inhomogeneities in the ART phantom.