Book Description

Different types of radiation detectors are routinely used for the dosimetry of photon beams. Finite detector sizes have certain effects to the broadening of the measured beam penumbra. The problem is more important in small field measurement, such as stereotactic radiosurgery, small beamlet IMRT, etc. The dosimetry associated with small fields is very difficult because of the steep dose gradients and the lack of lateral electronic equilibrium conditions that complicate the interpretation of the dose measurement. Many Researchers have investigated this problem from different points of view utilizing, for example, extrapolation method, analytical method. But their studies were all measurements based. In this study, we investigated the problem using Monte Carlo simulation method. Compared with practical measurements, the advantages of using Monte Carlo simulation are: 1. Simulation can be performed in a scenario where radiation dosimetry is technically difficult or even impossible to accomplish; 2. Possible systematic errors, e.g., setup errors, reading errors, can be eliminated; 3. Simulation of radiation detectors which are not readily available allowed the study of a wider range of detector sizes. In this study we used Monte Carlo methods to develop and apply detector response functions (DRFs) for three types of clinically available radiation detectors and two theoretical detectors. Detector response functions were determined by deconvolving known values of input (simulated true data from Monte Carlo simulation) and output (simulated empirical data from Monte Carlo simulation or empirical data from radiation dosimetry). Deconvolved detector response functions were applied to typical stereotactic radiosurgery fields to obtain the true beam profile. This application was then benchmarked by both Monte Carlo simulation method and dosimetry methods, which include diode dosimetry, radiographic film dosimetry, and Gafchromic film dosimetry. The results of this research demonstrate: 1. Detector response function of cylindrical detectors can be approximately represented as a Gaussian distribution dependent upon the radius of the detector; 2. Deconvolution method can create a more realistic beam profile by reducing the detector size effect, however it can not completely remove this effect limited by the inaccuracy derived from the Fourier transform-based nature of this procedure; 3. Diode dosimetry and Gafchromic film dosimetry both yield satisfactory beam profiles in small field relative measurements and are the preferred measurement techniques.