Project website: https://www.cancerprediction.org/.
The models created and used for this project are available in this repository.
As of May 2023, I have:
- Won the Rhode Island Science and Engineering Fair
- Become a Regeneron International Science and Engineering Fair (ISEF) Finalist
This project will primarily utilize the Python libraries Scikit-learn, TensorFlow, and XGBoost. The goal is for these models to accurately predict whether or not patients have various cancers such as colorectal, pancreatic, breast, etc, based on basic health features such as age, height, sex, etc. Data is provided by the National Cancer Institute's Cancer Data Access System, and will NOT be available in this repository.
This is an independent project for ISEF 2023. As of current, only colorectal data is being used.
Science Fair Poster
XGBoost Confusion Matrix Graphical (used)
XGBoost Confusion Matrix Numerical (used)
Neural Network Graphs (unused)
If I decide to revisit this project, I will:
- Integrate Large Language Models (LLMs) into the post-result recommendations
- Work with a medical professional
- Utilize various other deep-learning techniques for the prediction models
- Grow the website to include other cancers