GE Healthcare GE EK-Pro Arrhythmia Algorithm - Manual (Page 2)

How are arrhythmias monitored?
The GE EK-Pro algorithm processing can be represented by
three major phases: 1) Event Detection and Correlation,
2) Event Measurement and Classification, and 3) Arrhythmia
Detection. Continuous correlation, incremental template
updating and contextual analysis are processing techniques
used by the GE EK-Pro algorithm. Continuous correlation
attempts to find the best match between each incoming
complex and the set of stored (learned) templates. Incremental
template updating allows information from each beat,
which correlates over time, to be reflected in the associated
template. Contextual analysis uses information from
neighboring QRS complexes along with existing template
measurements to make the best possible decision regarding
the beat’s origin (e.g., early, wide). Additionally, the GE EK-Pro
algorithm also monitors ST changes of up to 12 leads.
For ST monitoring details, please consult the appropriate
user manual.
Event Detection and Correlation filters the incoming signal
before detecting the incoming events. As the events are
detected, incremental template updating is applied on the
signal. Here also the noise analysis is done in order to decide
whether if any of the incoming channels are too noisy to
be analyzed. Each detected QRS complex is compared to
previously detected QRS complexes and the templates are
updated accordingly, or a new template is created and the
beat is added on the beat list.
Event Measurement and Classification task uses the beat
list to define the timing, for example, the temporal information
of the patient rhythm. The algorithm can classify the beat in
question by using the information from the preceding and
following beats. A beat is classified as normal, ventricular or
artifact and, in combination of the decision of the past beat
template, the final label is given and the heart rate is also
updated accordingly.
Arrhythmia Detection uses the beat list to get information in
run lengths of normal and ventricular beats. Based on this, R-R
interval information and heart rate, the algorithm concludes
whether there are any arrhythmias present. The criteria for
calling an arrhythmia are listed in the table at the end of this
document. There is a separate analysis to detect the presence
of ventricular fibrillation.
The Advanced ECG tab provides the user with
options to adjust the ECG settings. Examples of
options include Pacemaker Detection, Relearning
QRS, and switching between multi-lead and single
lead analysis.
Practical aspects in bedside
arrhythmia monitoring
Signal quality
Careful skin preparation and the use of high-quality electrodes
are key to ensuring a good signal in ECG and especially
arrhythmia monitoring. A good signal helps ensure accurate
arrhythmia detection and helps decrease the number of false
alarms. In case there is an artifact in one of the analyzed
leads, the GE EK-Pro algorithm gives a message on noisy ECG
and, if the condition continues, the algorithm finally goes into
suspend mode. This is communicated by “Arrhyhtmia paused”
or “Arrhy suspend” message, depending on the GE monitor in
use. Artifacts can be a result of motion or poor skin-electrode
contact.
Relearning
When the morphology of the patient’s ECG changes
considerably (e.g. due to change of electrodes/electode
locations), relearning should be started manually. This can be
done in the ECG menu by selecting Relearn QRS.
Patients with pacemakers
Pacemaker detection needs to be activated when patients
with pacemakers are monitored. Pacemaker detection
is always on with the E-PSM or E-PRESTN module family.
Pacemaker detection needs to be activated when the
CARESCAPE
™ Patient Data Module or Tram ™ is used.
EK-Pro_DOC0783131.indd 2 11.11.2010 15.24