New specific absorbed fractions for annihilation radiation as a step towards a more individual dosimetry in nuclear diagnostics

In order to establish a more individual dosimetry for treatments involving radiopharmaceuticals like nuclear diagnostics and targeted radionuclide therapy, this project aims to calculate Specific Absorbed Fractions (SAFs) for annihilation radiation over a broad range of physiognomies. For the estimation of the SAFs varying with different physiognomic parameters, the study uses the Monte Carlo N-Particle Transport Code (MCNP) which during the last decades proved to be a powerful and reliable tool for dose calculations in human tissue. The human phantoms are generated from an additional software called The Body Builder which enables the user to generate a phantom according to his specifications. The phantoms generated by The Body Builder are based on the Christy/Eckermann phantoms with a first interpolation made on their coefficients and a subsequent transformation into the equivalent MCNP surface representation. The spectrum of age groups ranges from 1 to 21 years, covering both sexes. As source organs we chose the biokinetical compartments of 18 F-fluorodeoxyglucose (FDG) used in the model of Hays and Segall. A homogeneous distribution of the activity within the source organ is assumed, which means that the annihilation photons of 511 keV are considered to be emitted from an isotropic and homogenous volume source. In order to get good statistics the number of photons simulated ranged between 2 and 20 million, depending on size and distance between source and target organs. Results. Considering all possible combinations between source and target organs, representing 21 age groups of both sexes, one gets more than 30 000 SAFs. The work is still in progress during the preparation of this paper, but first results are already presented here. The relative error of the Monte Carlo simulation for every SAF mentioned below does not exceed 1,1%. An additional uncertainty factor of approximately 0,5% is introduced by using the p-mode of MCNP. The results of the different age groups are compared with the SAFs from the Christy/Eckermann series. This is done by linear interpolation between the two closest age groups from the Christy/Eckermann series (e.g. 5 and 10 for an 8 year old), a procedure often applied in clinical practice. These values are then compared to the ones calculated in the simulations described above. One has to mention that the Christy/Eckermann values refer to the energy of 500 keV, a value of slightly difference to the actual value of the annihilation radiation. Concentrating at the interesting age groups of 2-4, 6-9 and 10-14, where previously no explicit calculations existed one can now take a look at the different cases. For the cases where all the interpolated Christy/Eckermann values are higher and therefore represent a more conservative estimation in the sense of dose assessment, the new SAFs are smaller in a range from 1,5% to 8,9%, being 5% on the average. For the combinations where all the newly calculated SAFs are higher, the study so far bears an average increase of the SAF of 3,2% ranging from 0,5 to 6%. In total one can say that the new calculated SAFs differ from the interpolated Christy Eckermann values from -9% to +17%. However, 92% of the values lie between ±9%. All data mentioned above result from combinations of source and target organs both lying in the trunk and therefore being very close to each other