Department of Software Technology
Vienna University of Technology
Time-oriented Analysis of High-Frequency Data in ICU Monitoring
Abstract:
Interpretation
of high-frequency data (sampled every second) in Intensive Care Units (ICUs) requires a time-oriented analysis resulting in a high-level abstraction. The paper presents analysis methods of such data to abstract qualitative descriptions over a period of time. The methods are sensitive to the context and the abstracted descriptions cover both the current situation and the time course of the monitored parameters. The aim of the time-oriented analysis is to arrive at unified qualitative values or patterns, comprehensible to health care providers and easy to use for decision support.
The methods are guided by the expectation of skilled physicians about the physiological course of parameter development. They incorporate knowledge about data points, data intervals, and expected qualitative trend patterns. The applicability and usefulness are illustrated by examples from VIE-VENT, an open-loop knowledge-based monitoring and therapy planning system for artificially ventilated newborn infants.
Up
Comments: rauber@ifs.tuwien.ac.at