← Back to Explore

Towards Personalized Medicine: Using Machine Learning to Predict Immunotherapy Effectiveness on Heterogenous Tumors in silico

JSHS · 2023

Overview

Personalized medicine has the potential to revolutionize patient care using targeted therapeutics. Within the field of personalized medicine, simulation of tissue-scale biological systems has shown great potential in analyzing the effects of therapeutics. PhysiCell, an open-source agent-based modeling simulation platform, has the capability to model interactions of immune and cancer cells. However, PhysiCell simulations are time-consuming due to the millions of interacting cells. Machine learning presents a promising solution to predict the efficacy of a cancer treatment without running a PhysiCell simulation. Using PhysiCell, a training data set composed of 2000 immunosurveillance simulations was created with various parameters values. The simulations were run with two separate tumor morphologies based on histopathological images of pancreatic adenocarcinoma. Training data was used to create both a random forest classifier and a random forest regressor for each morphology. These trees predict the fraction of dead cancer cells based on parameter values and rank the parameter importance in predicting the fraction of dead cancer cells. The classifier predicted the fraction of dead cancer cells with accuracies of 70.91% and 84.55% for Morphologies One and Two, respectively. The regressor predicted the fraction of dead cancer cells with accuracies of 69.39% and 84.55% for Morphologies One and Two, respectively. The parameter importance rankings for regression and classification showed similarities within each morphology, while the parameters of importance varied between the two morphologies. Also, the regressor generates a prediction in under 10 seconds for a parameter set, while a PhysiCell simulation can take up to 40 minutes.

Competition history

  • JSHS 2023 Category not listed

Resources

Related projects

Closest projects by meaning, across every fair and year in the corpus.

Source: Junior Science and Humanities Symposium

Save projects to your library

Sign in with Google to keep track of projects you find interesting, organized into folders. An account also raises your daily allowance for “Has this been done?”, and lets you create a key for the MCP server with a much higher limit than anonymous use. Browsing stays public.

Continue with Google