Information Technology Reference
In-Depth Information
Chapter 15
Intelligent Adaptive Monitoring for Cardiac
Surveillance
Rene Quiniou
1
, Lucie Callens
2
, Guy Carrault
3
, Marie-Odile Cordier
4
, Elisa Fromont
5
,
Philippe Mabo
6
,andFran¸ois Portet
7
1
INRIA/IRISA Rennes, France
rene.quiniou@irisa.fr
2
INRIA/IRISA Rennes, France
lucie.callens@irisa.fr
3
LTSI, University of Rennes 1, France
guy.carrault@univ-rennes1.fr
4
IRISA, University of Rennes 1, France
marie-odile.cordier@irisa.fr
5
University of Lyon, University of Saint Etienne, France
Elisa.Fromont@univ-st-etienne.fr
6
Dept. of Cardiology, CHU-University of Rennes 1, France
philippe.mabo@univ-rennes1.fr
7
IMAG, Grenoble, France
francois.portet@imag.fr
Abstract.
Monitoring patients in intensive care units is a critical task. Simple
condition detection is generally insufficient to diagnose a patient and may gen-
erate many false alarms to the clinician operator. Deeper knowledge is needed to
discriminate among the flow of alarms those that necessitate urgent therapeutic
action. Overall, it is important to take into account the monitoring context: sensor
and signal context (are the signal data noisy or clear?), the patient condition (is
the state of the patient evolving or stable? is the condition of the patient critical
or safe?), the environmental context (do the external conditions influence the pa-
tient condition or not?). To achieve the best surveillance as possible, we propose
an intelligent monitoring system that makes use of several artificial intelligence
techniques: artificial neural networks for signal processing and temporal abstrac-
tion, temporal reasoning, model based diagnosis, decision rule based system for
adaptivity and machine learning for knowledge acquisition. To tackle the diffi-
culty of taking context change into account, we introduce a pilot aiming at adapt-
ing the system behavior by reconfiguring or tuning the parameters of the system
modules. A prototype has been implemented and is currently experimented and
evaluated. Some results, showing the benefits of the approach, are given.
1
Introduction
Monitoring means to process incoming data (signals) recorded by sensors in order to
recognize alarming conditions. Such devices may generate alarms in huge volume that
can overwhelm an operator who has to validate the alarms and take therapeutic actions.
Search WWH ::
Custom Search