Database Reference
In-Depth Information
capable of producing alarms and has reporting capabilities though a web
interface and through event notification mechanisms.
The Online Healthcare Analytics infrastructure, which is also known
as Artemis [27], is a programmable framework for real-time analysis
of intensive care sensor data leveraging the IBM InfoSphere Streams
(Streams) real-time high-performance stream analysis engine. OHA in-
terfaces Streams with an open set of data collection systems (e.g., Excel
Medical Electronics BedMasterEX system, the CapsuleTech data col-
lection system), and leverages different data mining technologies and
machine learning algorithms for the generation of models for prediction
of the onset of complications in intensive care. OHA leverages Streams
interface with well known analytic softwares such as SPSS, SAS and R to
provide data mining capabilities. Models learned with these data mining
systems can be scored in real-time, thus giving the analyst-physician the
ability to test clinical hypotheses prospectively. This analytical loop is
abstracted in
Figure 14.2
. It constitutes a general architecture for sensor
data mining applications leveraging both at rest analytics for modeling
andinmotionanalyticsforthescoringofmodelsinreal-time. The
OHA system has been in use in live environments for the monitoring of
neonates [27]. Its exploration capabilities are also used for the mining
of sensor data for the early detection of complications in neurological
ICUs [29].
Figure 14.2.
A generic architecture for sensor data mining systems
