Abstract:Medical histories of patients can predict a patient's immediate future. While most studies propose to predict survival from vital signs and hospital tests within one episode of care, we carried out selective feature engineering from longitudinal medical records in this study to develop a dataset with derived features. We thereafter trained multiple machine learning models for the binary prediction of whether an episode of care will culminate in death among patients suspected of bloodstream infections. The machine learning classifier performance is evaluated and compared and the feature importance impacting the model output is explored. The extreme gradient boosting model achieved the best performance for predicting death in the next hospital episode with an accuracy of 92%. Age at the time of the first visit, length of history, and information related to recent episodes were the most critical features.

Predicting In-Hospital Death from Derived EHR Trajectory Features