Early Detection of Sepsis using Ensemblers

TL;DR

This paper presents a novel ensemble-based methodology for early sepsis detection using hourly patient records, achieving high accuracy and utility scores through imputation and validation techniques.

Contribution

It introduces a weak ensembler approach combined with imputation for early sepsis detection in large medical datasets, which is a new application in this context.

Findings

Accuracy of 93.45% achieved

Utility score of 0.271 obtained

Validated with 3-fold cross-validation

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 ensembler technique to analyze these medical records and 3-fold validation, a model is created and validated internally. The model achieved an accuracy of 93.45% and a utility score of 0.271. The utility score as defined by the organizers takes into account true positives, negatives and false alarms.