This project will focus on the outcome of patients with sepsis in an intensive care unit (ICUs) setting. Particular outcomes of interest include: cardiac dysfunction, septic shock, acute kidney injury, and liver failure. We will identify physiological factors that lead to risk of organ failure in sepsis using the MIMIC III database. From there, we will characterize septic patients with different outcomes. We will then use this labeled dataset to build predictive models using machine learning techniques such as logistic regression and neural networks. The overall goal of the project is to explore the feasibility of implementing an early warning system to alert clinicians about patients at high risk of developing organ dysfunction.

Through the SuperUROP project, I hope to gain more experience working on a long-term, impactful research project. I am excited to apply the machine-learning techniques I have learned in my coursework to a real clinical dataset. With this project, I will use my knowledge to help people while improving my own skills.