Authors: Antoni Goldstein Michał Kardaś Piotr Kowalkowski Magdalena Molenda Łukasz Piekarski
Project done as a Bachelor thesis in Computer at University of Warsaw. Continuation of the work of the previous group accessible here tumor-ca and here EMR6-Ro.
Existing research provides a numerical method based on cellular automata to simulate growth of EMT6/Ro tumor spheroid under varying radiotherapy treatment protocols. As the space of possible protocols is very vast, the model has been used to conduct a heuristic genetic algorithm search for an optimal dosage and timing of irradiation. However, current imple- mentation of the simulation makes the search computationally costly, which limits the extent of explored solutions. We have created multiple datasets with 200 000 protocols each and tested with various machine learning algorithms. Thanks to the work of the previous group we had been able to use GPU to speed up the process and for every simulation we have conducted 100 simulations and took the mean of the results to predict expected value.
Build tumor-simulation executable. In project's main folder run:
mkdir build
cd build
cmake ..
make
Usage: Create protocols
./nasze-ca/build/protocol_generator
Run simulations using gpu
sbatch nasze-ca/gpu-gen.sl
Results will be in
nasze-ca/data/old_data/results/protocol_results_<number>.csv.