Skip to content

Machine learning model to predict heart transplant failure and success using XGBoost algorithm and SMOTE/ENN to balance the dataset.

Notifications You must be signed in to change notification settings

trevor-leach803/Heart_Transplant_Model

Repository files navigation

Heart_Transplant_Model

This project is done in partial conjunction with the Medical University of South Carolina (MUSC).

This project makes use of United Network for Organ Sharing data on heart transplants in the United States. Since this project has been worked on by MUSC, the data included in this repo is not original data. The data here has already been edited down from the original dataset and altered to meet the needs of this project.

The notebook here walks through the steps I took to reach the ideal model performance. The best model reached with this work so far is Model 5, which used the XGBoost algorithm and SMOTEENN to balance the dataset.

Additionally, this repo contains the poster presented at the College of Charleston EXPO on April 13, 2023. The final presentation for my machine learning graduate school class is also included alongside the final paper coded in LaTeX in IEEE format.

About

Machine learning model to predict heart transplant failure and success using XGBoost algorithm and SMOTE/ENN to balance the dataset.

Topics

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published