Introduction of a hybrid approach based on statistical shape model and Adaptive Neural Fuzzy Inference System (ANFIS) to assess dosimetry uncertainty: A Monte Carlo study.
Journal:
Computers in biology and medicine
PMID:
40068491
Abstract
The increasing use of ionizing radiation has raised concerns about adverse and long-term health risks for individuals. Therefore, to evaluate the range of risks and protection against ionizing radiation, it is necessary to assess the dosimetry calculation uncertainty of the absorbed dose of organs and tissues in the body. On the other hand, absorbed dose calculation with low computational load plays a noted role in dosimetry studies. Considering the Monte Carlo simulation's time-consuming and high computational cost, we present a novel model-based organ dosimetry for uncertainty evaluation. We attempt to model and estimate the organ-absorbed dose for lung organ size by combining computational phantoms and ANFIS. Two input variables were used, including variations in lung size and photon energy. The results showed that the proposed hybrid approach increased the speed of evaluation of the uncertainty of dosimetry calculations. The promising results of the hybrid approach demonstrate that it can be a suitable alternative to the time-consuming conventional methods of dosimetry calculations in dosimetry calculations, which will lead to the development of a rapid and reliable tool for organ dose estimation in dosimetry applications in the future.