Early Detection of Sepsis Using Ensemblers

Abstract

This paper describes a methodology to detect sepsis ahead of time by analyzing hourly patient records. The Physionet 2019 challenge consists of medical records of over 40,000 patients. Using imputation and weak ensem- bler technique to analyze these medical records and 3-fold validation, a model is created and validated internally. On a hidden test data set maintained by the organizers, the model obtained a utility score of 0.192. The utility score as defined by the organizers takes into account true positives, negatives and false alarms. Our team was Team Tesseract and our overall ranking was 49 out of 79 officially ranked entries.