A rule-based approach for the prediction of acute hypotensive episodes

Abstract

The 2009 Physionet Computers in Cardiology Challenge is to develop an automatic technique for the prediction of acute hypotensive episodes. Such episodes if left untreated can lead to organ failure and death. However, acute hypotensive episodes can be treated depending on its cause, and if diagnosed in time. This study uses a rule-based approach for the prediction of the acute hypotensive episodes. The physionet challenge is divided into two parts. The first is the ability to distinguish between patients who have experienced acute hypotensive episodes and patients who do not. The second part of the challenge is to predict acute hypotensive episodes. The technique uses the mean arterial blood pressure signal as an indicator for predicting AHE. The records used for this challenge are from the MIMIC II database. The challenge dataset is divided into 60 records for training, 10 records for testing the first part of the challenge, and 40 records for testing the second part of the challenge. The score is calculated as the fraction of correct classifications. The final results on the challenge are 1 and 0.875 for part 1 andpart 2, respectively.