Loading...

Modeling of Cancer Progression by Using Evolutionary Game Theory

Malekian Boroujeni, Negin | 2015

2379 Viewed
  1. Type of Document: M.Sc. Thesis
  2. Language: Farsi
  3. Document No: 47998 (19)
  4. University: Sharif University of Technology
  5. Department: Computer Engineering
  6. Advisor(s): Habibi, Jafar
  7. Abstract:
  8. Mathematical modeling and computer simulation are powerful tools to help cancer research because they provide a good insight of cancer progression and an efficient framework to test biological hypothesis. Most of the previous studies ignored many intracellular communication between tumor cells. Gap junction is one of the interaction ways between tumor cells which plays a crucial role in cancer progression. In the first phase of this thesis, a model of intracellular communication through gap junction using evolutionary game theory scoring is proposed. In the second phase of this thesis, cancer is modeled by Markov decision process (MDP). Given that considering nutrient level constant is one of the main weaknesses of the literature (cancer modeling by using Markov decision process), we solve this problem by modeling nutrient diffusion using a partial differential equation. The variables of this equation are continuous. Thus, a main challenge in the second phase is dealing with the continuous state space in the Markov decision process. We address this challenge by using a function approximation method to create features from the state space. Then, presented models are applied to ductal carcinoma in situ (DCIS), a type of breast cancer, and simulated. The results of the first phase show the role of gap junction communication in reducing cancer progression and the results of the second phase show the potential of multi-agent techniques for cancer modeling. Finally, NASA-STD-7009 is considered as a basis to evaluate our model
  9. Keywords:
  10. Reinforcement Learning ; Multi-Agent Reinforcement Learning ; Markov Decision Making ; Breast Cancer ; Evolutionary Games Theory ; Ductal Carcinoma Insitu (DCIS) ; Cancer Modeling

 Digital Object List

 Bookmark

...see more