Acute hypotensive episodes (AHE) is a critical event that can lead to irreversible organ damage and death in intensive care units (ICU). The goal of the 10 th annual PhysioNet/Computers in Cardiology Challenge is to predict which ICU patients will experience AHE within a forecast window of one hour.
In tackling this problem, most of the previous studies extract their features for AHE prediction from the time history of MAP, diastolic ABP and systolic ABP. In contrast, by exploring the interaction within the cardiovascular system, this work employs frequency domain approach. Toward this goal, this work proposes two feature sets: degree of concentration and energy from the spectrogram of the ECG and ABP signals. The mulstiscale entropy of these features have also been studied. The effectiveness of these features is statically investigated by comparing their means between the AHE and non AHI patient groups.
Identifer | oai:union.ndltd.org:NSYSU/oai:NSYSU:etd-0718112-163846 |
Date | 18 July 2012 |
Creators | Huang, Shen-Tung |
Contributors | Pei-Chung Chen, Chen-Wen Yen, Jiann-Der Lee |
Publisher | NSYSU |
Source Sets | NSYSU Electronic Thesis and Dissertation Archive |
Language | Cholon |
Detected Language | English |
Type | text |
Format | application/pdf |
Source | http://etd.lib.nsysu.edu.tw/ETD-db/ETD-search/view_etd?URN=etd-0718112-163846 |
Rights | unrestricted, Copyright information available at source archive |
Page generated in 0.0531 seconds