Design and simulation of a Radiotherapy System for Breast Cancer Treatment

The first Mexican RF LINAC for cancer treatment.

This research project consists of the design of an electron LINAC to treat breast cancer in Mexico. The aim of the project is to make a first design of the main elements that make up this accelerator and optimizing them to achieve a compact and energy efficient design. The project began by determining the type of accelerator and the type of particles that would be more appropriate for the breast cancer treatment considering the situation of medical equipment in Mexico. The main elements of the accelerator were identified and we began their design and simulation. The project consists of 3 main areas: the acceleration system, the radiation model and the clinical/medical area. 

Technologies used

The design of the LINAC has been almost entirely made  with simulating software. For designing the beamline and tracking the particles along the accelerator we used MADX. This is a software that is useful for calculating TWISS parameters and simulating (tracking) how a particle distribution evolves along an accelerating structure that can include accelerating cavities, focusing elements, bending magnets, etc. For studying the beam as it leaves the accelerator and it travels to and through different materials (first water and then a more realistic breast tissue) we used Geant4. This is a software that has been useful for the entire radiation model because we use it to simulate how energy is deposited in the tissue at different depths. We have complemented this study with simulations done in Python.


We have faced two kinds of barriers for this project: an intellectual and a technological one. We are still undergraduate students working in a project of particle accelerators, which is a very complex and advanced topic for . The second challenge has been learning, understanding and manipulating the softwares needed to develop the project in a meaningful way. We went through a very steep learning curve with our two main softwares, Geant4 and MADX, because they are not very user friendly and at least for MADX there is little to no information about it. We have faced the challenge of implementing new and complicated topics into algorithms in softwares we have little experience with. 


For many of the people joining the hackathon it was great to be able to contribute to an existing project, the challenge of finding a problem and solving it was intellectually stimulating as well as giving us a bit more mobility than other projects might have provided.


We learned a lot about biophysics and got to practice previous skills in python and problem solving. Also we realized the complexity of the acceleration system, that demands adjacent components to work efficiently. 

Next steps

The immediate next steps include finalizing the project at this first stage of design. We still need to make optimizations for the size of the accelerator as well as the amount of particle losses. Another important next step  is successfully connecting the accelerator simulation to the radiation simulation. This will be done by taking the output of MADX, the distribution of particles (their position, momentum and energy), and using it as the input distribution for the Geant4 code. This will mean both simulations are connected end to end. 

Some medium to long term next steps include making a collaboration with the breast cancer treatment center of the Zambrano Hellion Hospital. 

One-minute video