This repository contains the code for the course project titled: "Optimal Treatment Policy Formulation Using Causal Trees - A Case Study on Critical Care Patients Receiving Invasive Mechanical Ventilation".
Project Summary:
Personalized healthcare (PHC) aims to tailor optimal treatments for patients based on their heterogeneity. Causal inference plays a significant role in supporting PHC by providing estimation tools for subpopulation based heterogeneous treatment effect (HTE) estimation. In this study, our goal is to leverage a unique causal inference tool - causal trees to estimate HTEs for critical care patients who need ventilation support. Precisely, we aim to identify the optimal treatment strategy - liberal vs. conservative oxygenation and important covariates for different groups of ICU patients based on their HTE estimates using multiple types of causal trees. From the experimental findings, we can deduct a set of decision rules which further needs to be validated by taking the opinions of clinical researchers and domain experts.