RECONSTRUCCIÓN DEL ESPECTRO DE ENERGÍA DE FOTONES A PARTIR DE CURVAS CLÍNICAS DE DOSIS PORCENTUAL EN PROFUNDIDAD MEDIANTE REGULARIZACIÓN DE TIKHONOV Y GENERALIZED SIMULATED ANNEALING

The primary objective of this study is to reconstruct the energy spectrum from three linear accelerators (LINACs) using experimental measurements of the percentage depth dose (PDD) curve. The experimentally obtained percentage depth dose curves were used to solve the Fredholm integral equation. The photon beam spectra are related to radiation doses through a Fredholm integral equation, utilizing the generalized simulated annealing optimization method. The resulting spectrum was used to simulate an irradiation reference condition as recommended by TRS-398. The Monte Carlo codes PENELOPE and TOPAS were employed to create the simulation scenario under reference conditions (10 x 10 cm² field size, 100 cm SSD, and 30 x 30 x 30 cm³ water phantom) for 6 MV photon beams. The calculated spectra from the three LINACs demonstrated a remarkable level of concordance, achieving up to 99% agreement. The validation of the reconstructed spectrum was carried out by comparing it with the PDD and beam profile curves, revealing a highly favorable correspondence in their behavior. A comprehensive analysis compared the experimentally acquired PDDs with those simulated using the reconstructed spectrum. Parameters such as the entrance dose and were derived from the PDD curves for evaluation. Upon conducting a thorough comparison of these parameters with the experimental dataset, noticeable deviations of 10 % (entrance dose), and 3 % ( ). Beam profile comparisons across field size dimensions revealed differences ranging from 0.5% to 5.3%. The present study encompassed the reconstruction of the photon beam spectrum originating from LINACs, revealing a noteworthy level of agreement among them. The validation of the Fredholm integral equation by utilizing two simulation codes, as facilitated by the analysis of the PDD and beam profile curves, revealed substantial disparities within the region leading up to the build-up point. This reconstructed spectrum holds considerable potential for simulation scenarios within radiotherapy applications. This significance is particularly underscored by the challenges associated with acquiring comprehensive data from manufacturers of LINACs, which impedes access to crucial information regarding the constituents of these accelerators.