The three-dimensional internal dosimetry software package, 3D-ID

The standard formalism for patient dosimetry was developed by the Medical Internal Radiation Dose (MIRD) Committee and is described in reference [1], and also reviewed in chapter 4 of this book. As indicated above, to simplify absorbed dose calculations, the MIRD Committee developed ‘S factor’ tables and an associated procedure for their use [2]. This procedure has been implemented in a software package, MIRDOSE3 [3]. To generate tables of S factors for different radionuclides and source-target organ combinations, a standard model of human anatomy was adopted in which organ position, dimensions, and composition were mathematically defined. The radioactivity was assumed to be uniformly distributed throughout each source organ and the S factors were defined as the mean absorbed dose to a target organ per unit cumulated activity in a source organ. Since the position and size of tumours may vary within the body and since a standard model of human anatomy was adopted for generating the S factor tables, tumours are not included in the published tables. A number of approaches have been developed for estimating the absorbed dose to tumours and the dose contribution from tumours to normal organs. The simplest approximation is made by assuming that all electrons are deposited locally and that the relative contribution to the tumour-absorbed dose from photons is negligible. Alternatively, the fraction of electron energy absorbed may be considered assuming the tumour can be modelled as a sphere [4, 5]. Using tables of photon-absorbed fraction to spheres or ellipsoids, the photon self-dose may be added by assuming that the tumour is a sphere or ellipsoid [6]. If this assumption is made, the photon dose to and from normal organs may also be calculated by placing the idealized tumour geometry in a defined posi-calculations If a point-kernel convolution technique is used in estimating absorbed dose, the true tumour and normal tissue geometry as well as the activity distribution may be taken into account to yield a spatial absorbed dose or dose-rate distribution [9-20]. Tissue composition and density variations are not easily accounted for using point-kernel techniques. To account for these, Monte Carlo techniques are needed to estimate absorbed dose [21-32].