ABSTRACT

Radiation therapy is one of the most widespread treatments applied to certain types of cancer patients. However, the effectiveness of this type of treatment destructing the cancer cells is related to the side effects of radiation possibility on the surrounding healthy tissues. The risk of damage at healthy cells depends primarily on the direction of the emitted beam by the radiation unit and the intensity of the radiation received by the patient. 
Technological advances are enabling optimize treatments, decreasing administered doses and its undesirable effects, but one of the main problems in these systems dose calculation is the calculation accuracy of the algorithms in the presence of tissues with different densities, as well as the exact knowledge of the spectrum emitted by medical linear accelerators. 
One of the tools used in this field is the Monte Carlo method instead of deterministic calculation procedures. This method is a calculation technique that allows, among other applications, simulate the effect of radiation used in cancer therapy and other similar conditions. 
The work presented here tries to demonstrate the possibility of using Monte Carlo simulations in radiotherapy treatment planning, improving the efficiency in dose distribution calculation in a particular environment compared to traditional systems. In addition, the work also tries to validate the use of simulations in other related applications such as the reconstruction of photonic spectra. 
A simulation of this type involves modeling with realism the linear accelerator head geometry and defining the physical parameters governing the particles transport. Detailed knowledge of the spectrum emitted by the linear accelerator is also essential, since the dosimetric factors depend directly on the beam energy. An important part of this work has been focused on the spectrum reconstruction of a linear accelerator and its use in simulating the photon and electron transport during the radiotherapy unit operation. The developed procedure to characterize the beams generated in a linear accelerator irradiation is based on Hansen deconvolution algorithms using the simulated and measured depth dose curves in a water phantom. 
Comparisons between experimental measures and calculations results show that the developed algorithm in this work is a valid way to reconstruct the photon spectrum emitted by radiation therapy unit. The reconstruction process is explained in detail and rigorously throughout this work, in which experimental data will be provided to justify that this methodology is a reliable technique to achieve the purpose of reconstructing photonic spectra. 
The particles transport code used in the simulations of this thesis has been the Monte Carlo N-Particle Transport Code System (MCNP), fifth version, developed at the Alamos Laboratory, (United States) and chosen as one of the most accurate calculation programs in the field of neutrons, photons and electron transport simulations.