Philips SpO2 Algorithm: Enhanced Accuracy
Uploaded by
lehkedaPhilips SpO2 Algorithm: Enhanced Accuracy
Uploaded by
lehkedaPUBLISH
Philips Pulse Oximetry
A More Reliable Approach for Deriving
SpO2 and Pulse Rate
Application Note
Introduction
The attributes of a measurement that make it reliable are Philips Healthcare’ SpO2 Algorithm
its accuracy and its repeatability. Accuracy in the sense of
getting the right value, repeatability in the sense that the This patented algorithm from Philips still derives the
value is not significantly affected by “external” SpO2 concentration from the absorptions of red and
conditions. infra-red light, but it takes an entirely different approach
to getting a good, reliable, and clean signal.
When it comes to modern patient monitoring, these two
attributes are principally determined by the mathematical Instead of examining how the strength of the electrical
process, or algorithm, used to convert the electrical signal signals change over time, the Philips algorithm examines
from the transducer into meaningful wave forms and the strength of the different frequency components that
numerics on the screen. The main difficulties in getting make up the signals.
the algorithm right include getting a good signal and The most significant frequency component is the pulse
dealing with the “noise” added to the electrical signal by rate. Because of the way the blood pulses in the arteries,
external conditions. there are also components at multiples of the pulse rate.
Traditional SpO2 algorithms have consistently had The frequency components of a signal with no noise are
problems with the noise created by factors such as patient illustrated in Figure 1.
motion, including shivering, and in dealing with weak
signals (if the patient has low perfusion).
Philips Pulse Oximetry
Figure 1 Frequency Components without Finding the Pulse Rate Component
Noise The Philips algorithm takes a number of samples of the
electrical signal and uses them to derive the full spectrum
Signal of frequency components. It then applies a number of
Strength
complex criteria to select the component that is at the
pulse rate for both the red and infra-red signal. Among
the criteria that are taken into account are:
Is this a likely pulse rate for the type of patient (adult
or pediatric)?
Is this pulse rate close to the last 3 accepted pulse
Pulse 2x 3x
rates?
... frequency
Rate Pulse Pulse Are there frequency components at multiples of the
Rate Rate proposed pulse rate?
A “real” signal with noise is illustrated in Figure 2: Is there a good correlation between the red and
infra-red spectra at this frequency?
Figure 2 Frequency Components with Noise Is the resulting SpO2 value using the red and infra-red
frequency components valid and reasonable?
Signal
Strength Is the resulting SpO2 value close to the last 3 accepted
SpO2 values?
Accuracy and Freedom from False
Alarms
Once a frequency component has been selected, it is then
2x 3x ... frequency used for the derivation of the SpO2, the pulse rate and
Pulse Pulse Pulse the perfusion index numerics.
Rate Rate Rate
Noise is easily and reliably rejected because it fails to meet
most of the criteria (only some of which have been listed
here). Noise typically occurs at the wrong frequencies,
has the wrong signal strength and occurs at random,
changing frequencies.
The result is an increase in measurement accuracy
(particularly noticeable for patients with low perfusion),
but it also means a significant reduction in the number of
false alarms, an achievement that has been proven in
recent laboratory results.
©2014 Published March 2014, Edition 1
Koninklijke Philips N.V.
453564491211
All rights are reserved. Reproduction in whole or in part is prohibited without
the prior written consent of the copyright holder. *453564491211*
Philips Healthcare reserves the right to make changes in specifications or to
discontinue any product at any time without notice or obligation and will not
*1.0*
be liable for any consequences resulting from the use of this publication.
2
PUBLISH