Calculation of air kerma inside the radiation field of X-ray tube.

The air kinetic energy released per unit of mass (air kerma) quantity is of importance in both medicine and industry for evaluating the radiation field of X-ray tubes and also for calibrating the reference instruments. The air kerma spatial distribution inside the X-ray tube's radiation field is not uniform because of the heel effect phenomenon. In this paper, a combination of Monte Carlo N Particle eXtended (MCNPX) simulation code and artificial neural network (ANN) has been utilized to estimate the air kerma inside the radiation field of X-ray tube for a wide range of tube voltages (50–600 kV) which covers both medical and industrial applications. At first, an X-ray tube with an anode angle of 20° was simulated using MCNPX code. Next, in order to provide the required data for training the network, 1375 point detectors were placed in different positions inside the conical radiation field of the simulated X-ray tube and then the air kerma was calculated for tube voltages in the range of 50–600 kV with a step of 50 kV. The tube voltage and point detector's spherical coordinates including distance from target, tangential angle and polar angle were utilized as the 4 inputs and the air kerma was used as the output of the ANN. After training the ANN, the proposed ANN model could estimate the air kerma in every position inside the radiation field of X-ray tube for a wide range of voltages (50–600 kV) with a mean relative error (MRE) of less than 0.67%. • MCNPX and ANN were utilized to estimate the air kerma inside the radiation field of X-ray tube. • 1375 point detectors were placed inside the conical radiation field of the X-ray tube and then air kerma was calculated. • The tube voltage and point detector's spherical coordinates were utilized as the 4 inputs of the ANN. • The proposed ANN model could estimate the air kerma with a MRE of less than 0.67 %. [ABSTRACT FROM AUTHOR]