A Graphical User Interface For Diagnostic Radiology Dosimetry Using Monte Carlo (MCNP) Simulation

INTRODUCTION Monte Carlo (MC) methods have become the "gold standard" for assessing dose distribution, organ doses and effective dose associated with the exposure, both external and internal, of humans to ionizing radiation. Powerful and sophisticated tools for simulation of radiation transport (for example, MCNP, EGS4) are now widely available. However, these codes are designed for general-purpose use and consequently do not provide a simple user interface for specific dosimetry tasks. The application of MC codes to diagnostic radiology dosimetry is a 2-stage process. Firstly, an appropriate input must be generated to describe the configuration of the irradiation conditions in a code-specific format. This includes the patient, who is represented by a mathematical phantom, and the geometry/quality of the beam, which is determined by the technical settings (tube voltage and filtration) of the X-ray unit to be simulated. Finally, after running the MC program, the data written to an output file must be processed to extract and present the results. Manipulation of input and output text files is both time-consuming and error prone. Consequently, without some form of automation, the routine application of the MC methods, (for instance in hospital radiology departments), is difficult to achieve. We have developed a graphical user interface (XrayDose), to facilitate the use of the MC code, MCNP, for diagnostic radiology dosimetry. The program provides a user friendly interface for (i) modifying the MCNP input file, (ii) launching the MCNP program to simulate photon transport through the anthropomorphic mathematical phantom and (iii) processing the MCNP output file to yield a summary of the results (organ and effective dose). In this paper the development and characteristics of XrayDose are outlined. As part of the validation process, the results for 2 diagnostic radiology studies (Chest PA and Abdomen AP) using XrayDose are compared to literature data obtained with another MC code and phantom.