Building a COVID-19 Vulnerability Index

Dave DeCaprio; Joseph A Gartner; Thadeus Burgess; Sarthak Kothari; Shaayaan Sayed; Carol J McCall

Here, we create two variations of the models. The first is a model that leverages information similar to our logistic regression model. A nice feature of gradient boosted trees is that they are fairly robust against learning features that are eccentricities of the training data, but do not extend well to future data. As such, we allow full diagnosis histories to be leveraged within our simpler XGBoost model. In this approach, every category in the full CCSR is converted into an indicator feature, resulting in 559 features. Details about how to connect the full diagnosis history with the open-source model are provided with the open-source version of the model. We additionally built a model within the ClosedLoop platform. The ClosedLoop platform is a software system designed to enable rapid creation of machine learning models utilizing healthcare data. The full details of the platform are outside the bounds of this paper; however, using the platform allows us to leverage engineered features coming from peer-reviewed studies. Examples are social determinants of health and the Charlson Comorbidity Index [13] . We chose not to include these features within the open-source model, because the purpose of the open-source version is intended to be as accessible as possible for the greater healthcare data science community.