| Article ID | Journal | Published Year | Pages | File Type |
|---|---|---|---|---|
| 723345 | IFAC Proceedings Volumes | 2006 | 6 Pages |
Abstract
: This paper presents a methodology to automatically detect specific medical events occurring on patients from ICU high frequency physiological variables. The physiological variables are transformed into a succession of temporal episodes (trend) and residual by an adaptive trend extraction method. Event models using temporal episodes from several variables are proposed. Results obtained on 70 hours of recordings from 23 adult patients hospitalised in ICU assess the capacity of the method to recognize events.
Related Topics
Physical Sciences and Engineering
Engineering
Computational Mechanics
Authors
Sylvie Charbonnier,