ThesisPDF Available

False arrhythmia alarm suppression using ECG, ABP, and photoplethysmogram

Authors:

Abstract and Figures

Thesis (M. Eng.)--Massachusetts Institute of Technology, Dept. of Electrical Engineering and Computer Science, 2009. This electronic version was submitted by the student author. The certified thesis is available in the Institute Archives and Special Collections. Cataloged from student-submitted PDF version of thesis. Includes bibliographical references (p. 91-93). A signal quality assessment scheme for the photoplethysmogram waveform recorded by a pulse oximeter has been created. The signal quality algorithm uses statistical methods on time-series and spectral analysis to locate high-frequency segments of the photoplethysmogram waveform. A photoplethysmogram pulse onset detector has been implemented for heart rate estimation. Application of the signal quality metric and photoplethysmogram pulse onset detector are demonstrated in an algorithm which suppresses false electrocardiogram critical arrhythmia alarms issued by bedside monitors in hospital intensive care units. by Anagha Vishwas Deshmane. M.Eng.
Content may be subject to copyright.
False Arrhythmia Alarm Suppression Using ECG,
ABP, and Photoplethysmogram
by
Anagha Vishwas Deshmane
S.B., Massacusetts Institute of Technology (2008)
Submitted to the Department of Electrical Engineering and Computer
Science
in partial fulfillment of the requirements for the degree of
Master of Engineering in Electrical Engineering and Computer Science
at the
MASSACHUSETTS INSTITUTE OF TECHNOLOGY
September 2009
c
Massachusetts Institute of Technology 2009. All rights reserved.
Author ..............................................................
Department of Electrical Engineering and Computer Science
August 21, 2009
Certied by..........................................................
Dr. Roger G. Mark
Distinguished Professor in Health Science & Technology
M.I.T. Thesis Supervisor
Certied by..........................................................
Lauren J. Kessler
Charles Stark Draper Laboratory
VI-A Company Thesis Supervisor
Accepted by.........................................................
Dr. Christopher J. Terman
Chairman, Department Committee on Graduate Theses
2
False Arrhythmia Alarm Suppression Using ECG, ABP, and
Photoplethysmogram
by
Anagha Vishwas Deshmane
Submitted to the Department of Electrical Engineering and Computer Science
on August 21, 2009, in partial fulfillment of the
requirements for the degree of
Master of Engineering in Electrical Engineering and Computer Science
Abstract
A signal quality assessment scheme for the photoplethysmogram waveform recorded
by a pulse oximeter has been created. The signal quality algorithm uses statistical
methods on time-series and spectral analysis to locate high-frequency segments of
the photoplethysmogram waveform. A photoplethysmogram pulse onset detector has
been implemented for heart rate estimation. Application of the signal quality met-
ric and photoplethysmogram pulse onset detector are demonstrated in an algorithm
which suppresses false electrocardiogram critical arrhythmia alarms issued by bedside
monitors in hospital intensive care units.
M.I.T. Thesis Supervisor: Dr. Roger G. Mark
Title: Distinguished Professor in Health Science & Technology
VI-A Company Thesis Supervisor: Lauren J. Kessler
Affiliation: Charles Stark Draper Laboratory
3
4
Acknowledgments
I would like to thank Dr. Roger Mark and Dr. Gari Clifford for sharing their medical
insight, engineering expertice, and guidance on the general direction of this research.
Thanks to Omar Abdala for his discussions on Hjorth parameters, and Daniel Scott
and Mauricio Villarroel for their extensive computing knowledge and help with nav-
igating the MIMIC II database and related tools. Thanks to the members of the
Bioengineering Research Partnership for their feedback during group meetings. Last
but not least, I would like to express my deepest gratitude to my VI-A advisor, Lau-
ren Kessler, for several years of encouragement, excellent mentorship, and guidance
during the thesis-writing process.
Thanks to the members of the Lab for Computational Physiology at M.I.T. for
making my time in the lab enjoyable, and teaching me how to take a lunch break
and get addicted to coffee. Special thanks to the resident students, Tiffany Chen
and Shamim Nemati, and the visiting students, Violetta Monastario Bazin and Patti
Ordonez Rozo, for keeping me company during nights and weekends in the lab, and
providing endless hours of amusement.
Finally, I would like to thank my parents, Vishwas and Meera, and my sister,
Anisha, for a lifetime of inspiration, encouragement, and continued support of all my
endeavors.
This work was funded in part through the Draper Fellows program at the Charles
Stark Draper Laboratory, under contract numbers 22951-0001 and 23985-001. Pub-
lication of this thesis does not constitute approval by the Charles Stark Draper Lab-
oratory. It is published for the exchange and stimulation of ideas.
The work described here was also supported by Grant Number RO1-EB001659
from the National Institute of Biomedical Imaging and Bioengineering. The content
is solely the responsibility of the author and does not necessarily represent the official
views of the National Institute of Biomedical Imaging and Bioengineering or the
National Institutes of Health.
5
THIS PAGE INTENTIONALLY LEFT BLANK
6
Contents
1 Introduction 15
1.1 Motivation and Background . . . . . . . . . . . . . . . . . . . . . . . 15
1.2 The Photoplethysmogram Waveform . . . . . . . . . . . . . . . . . . 17
1.2.1 PulseOximetry.......................... 17
1.2.2 Waveform Morphology . . . . . . . . . . . . . . . . . . . . . . 23
1.2.3 Artifacts.............................. 23
1.3 OverviewofThesis ............................ 25
2 Signal Quality Assessment 27
2.1 PreviousWork .............................. 27
2.2 Modifications for Prototype Artifact Detector . . . . . . . . . . . . . 32
2.3 Adaptive Assessment of Signal quality . . . . . . . . . . . . . . . . . 35
2.3.1 Structure and Availability of Waveform Data . . . . . . . . . . 35
2.3.2 Preparation of Alarm Data . . . . . . . . . . . . . . . . . . . . 37
2.3.3 Preparation of Normal Sinus Rhythm Data . . . . . . . . . . . 39
2.3.4 Hjorth Parameter Assessment By Alarm Type . . . . . . . . . 41
2.3.5 Threshold Setting . . . . . . . . . . . . . . . . . . . . . . . . . 42
2.4 Use of pSQI ................................ 44
3 PPG Pulse Onset Detection 45
3.1 PreviousWork .............................. 45
3.2 aP P G: Photoplethysmogram Pulse Onset Detection . . . . . . . . . 47
3.3 aP P G Performance............................ 49
7
3.3.1 Data Acquisition, Pre-processing, and Evaluation Setup . . . . 49
3.3.2 Results............................... 52
3.3.3 Discussion of Limitations . . . . . . . . . . . . . . . . . . . . . 52
3.3.4 Future Work: Parameter Optimization and Testing . . . . . . 55
3.4 Use of aP P G ............................... 56
4 A New False ECG Alarm Suppression Framework Using the PPG
Waveform 59
4.1 Algorithm Architecture . . . . . . . . . . . . . . . . . . . . . . . . . . 59
4.1.1 Asystole Processing . . . . . . . . . . . . . . . . . . . . . . . . 61
4.1.2 Extreme Bradycardia Processing . . . . . . . . . . . . . . . . 62
4.1.3 Extreme Tachycardia Processing . . . . . . . . . . . . . . . . . 62
4.1.4 Ventricular Tachycardia Processing . . . . . . . . . . . . . . . 62
4.1.5 Ventricular Fibrillation Processing . . . . . . . . . . . . . . . . 62
4.2 Optimization of Signal Quality Thresholds . . . . . . . . . . . . . . . 63
4.3 Performance of PPG-Based False Alarm Suppression . . . . . . . . . 68
4.4 Limitations and Possible Improvements . . . . . . . . . . . . . . . . . 69
5 Conclusions 73
5.1 Summary ................................. 73
5.1.1 Contributions........................... 73
5.1.2 Evaluation and Limitations . . . . . . . . . . . . . . . . . . . 73
5.2 FutureWork................................ 75
5.2.1 pSQI Improvement........................ 75
5.2.2 aP P G Improvement ....................... 77
5.2.3 False Alarm Suppression Improvement . . . . . . . . . . . . . 78
5.2.4 Other applications . . . . . . . . . . . . . . . . . . . . . . . . 79
5.3 Extensibility................................ 79
A False ECG Alarm Suppression Using the ABP Waveform 81
A.1 Original Algorithm Architecture . . . . . . . . . . . . . . . . . . . . . 81
8
A.1.1 Asystole Processing . . . . . . . . . . . . . . . . . . . . . . . . 82
A.1.2 Extreme Bradycardia Processing . . . . . . . . . . . . . . . . 83
A.1.3 Extreme Tachycardia Processing . . . . . . . . . . . . . . . . . 83
A.1.4 Ventricular Tachycardia Processing . . . . . . . . . . . . . . . 83
A.1.5 Ventricular Fibrillation Processing . . . . . . . . . . . . . . . . 83
A.1.6 Performance on Unseen Data . . . . . . . . . . . . . . . . . . 84
A.1.7 Limitations ............................ 86
A.2 Modifications Made for Benchmarking . . . . . . . . . . . . . . . . . 87
A.2.1 Performance on Unseen Data . . . . . . . . . . . . . . . . . . 87
9
THIS PAGE INTENTIONALLY LEFT BLANK
10
List of Figures
1-1 Using the photoplethysmogram to corroborate ECG alarms . . . . . . 18
1-2 Absorption spectrum of hemoglobin species . . . . . . . . . . . . . . . 19
1-3 Light absorption waveform in inhomogenious tissue. . . . . . . . . . . 21
1-4 Empirical determination of %SpO2estimates based on light intensity
ratio, R.................................. 22
2-1 Hjorth parameter calculation for PPG segments at various heart rates 31
2-2 PPG artifact detection based on Hjorth parameters . . . . . . . . . . 34
2-3 Waveform data available with critical electrocardiogram alarms . . . 37
2-4 Box and whisker plot of mobility parameter (H1) distributions by alarm
type and condition (veracity) . . . . . . . . . . . . . . . . . . . . . . 42
2-5 Box and whisker plot of complexity parameter (H2) distributions by
alarm type and condition (veracity) . . . . . . . . . . . . . . . . . . . 43
3-1 Use of the Slope Sum Function to detect pulse onsets in the arterial
blood pressure waveform. Adapted from Figure 4 in [32]. . . . . . . . 47
3-2 PPG pulse onset detection by aP P G under conditions of normal sinus
rhythm, asystole, and bradycardia . . . . . . . . . . . . . . . . . . . . 50
3-3 PPG pulse onset detection by aP P G under conditions of tachycardia
and ventricular tachycardia . . . . . . . . . . . . . . . . . . . . . . . 51
4-1 False ECG Alarm Suppression Using the PPG Waveform . . . . . . . 61
4-2 Effect of ηl
1,ηu
1, and η2on true and false alarm suppression rates during
asystole .................................. 64
11
4-3 Effect of ηl
1,ηu
1, and η2on true and false alarm suppression rates during
extremebradycardia ........................... 65
4-4 Effect of ηl
1,ηu
1, and η2on true and false alarm suppression rates during
extremetachycardia ........................... 66
4-5 Effect of ηl
1,ηu
1, and η2on true and false alarm suppression rates during
ventricular tachycardia . . . . . . . . . . . . . . . . . . . . . . . . . . 67
A-1 False ECG Alarm Suppression Using the ABP Waveform . . . . . . . 82
12
List of Tables
2.1 Estimated hours of available waveform data . . . . . . . . . . . . . . 37
2.2 Annotated critical ECG arrhythmia alarms in gold standard database.
For example, there are 29 true asystole alarms, indicating that 1.2% of
all alarms in the database are true asystole alarms, and that 7.8% of
all asystole alarms in the data set are true. . . . . . . . . . . . . . . . 39
2.3 Annotated critical ECG arrhythmia alarms in Training Set. For ex-
ample, in the training set there are 29 true asystole alarms, indicating
that 1.3% of all alarms in the training set are true asystole alarms, and
that 8.3% of all asystole alarms in the training set are true. . . . . . . 40
2.4 Annotated critical ECG arrhythmia alarms in Test Set. For example,
in the test set there are 21 true asystole alarms, indicating that 1.2%
of all alarms in the test set are true asystole alarms, and that 7.2% of
all asystole alarms in the test set are true. . . . . . . . . . . . . . . . 40
2.5 Results of Kolmogorov-Smirnov tests for H1and H2during true and
false alarms to be sampled from different distributions . . . . . . . . . 44
2.6 Ranges of Hjorth parameter threshold settings tested for each alarm
type .................................... 44
3.1 Performance of aP P G on MIMIC I database . . . . . . . . . . . . . . 53
4.1 Windowing and thresholding parameters in merged PPG-based false
alarm suppression algorithm . . . . . . . . . . . . . . . . . . . . . . . 63
4.2 Optimal assignment of Hjorth parameter thresholds by alarm type us-
ingtrainingdata ............................. 65
13
4.3 Performance of PPG-based false alarm suppression algorithm . . . . . 68
A.1 Performance of ABP-based false alarm suppression algorithm reported
by Aboukhalil et al. ............................ 84
A.2 Performance of ABP-based false alarm suppression algorithm on new
MIMICIIdata .............................. 85
A.3 Windowing and thresholding parameters in merged ABP-based false
alarm suppression algorithm . . . . . . . . . . . . . . . . . . . . . . . 88
A.4 Performance of modified ABP-based false alarm suppression algorithm
onnewMIMICIIdata.......................... 88
14
Chapter 1
Introduction
1.1 Motivation and Background
Falsely issued alarms in intensive care units (ICUs) disrupt patients’ rest, drain hospi-
tal resources, and desensitize the hospital staff to potential emergency situations [2].
It has been estimated that 43% of life-threatening electrocardiogram (ECG) alarms
issued by bedside monitors are false, with some categories of alarm being as high
as 90% [1]. These false arrhythmia alarms are often triggered by noise and other
artifacts in the monitored ECG waveform, and can be suppressed in the presence of
other data which indicate that there are no critical abnormalities in cardiac function.
Such information can come from signals which are related to cardiac function but are
measured in a location remote to the heart and are therefore unlikely to exhibit the
same types of noise and artifacts as the ECG. Signals with pulsatile waveforms offer
the additional benefit of having features indicative of the cardiac cycle, which can be
later compared to timing and morphology of features in the ECG waveform.
Aboukhalil et al. have created an algorithmic framework which consults the arte-
rial blood pressure (ABP) waveform to corroborate critical ECG arrhythmia alarms
[1, 3]. If an ECG alarm is triggered, the algorithm checks the signal quality of the
simultaneously recorded ABP waveform. If this waveform is of poor quality, the ECG
alarm is accepted as true. If the ABP signal is of high quality, the algorithm checks
that the features extracted from the blood pressure waveform corroborate the condi-
15
tion which triggered the ECG alarm. The alarm is suppressed if the blood pressure
waveform does not exhibit features consistent with a cardiac arrhythmia.
The blood pressure signal quality assessment scheme in this framework, designed
by Sun et al. [26], uses a binary signal abnormality index, jSQI, to indicate if each
blood pressure beat is unsuitably noisy. The jSQI algorithm detects the onset of
each pulse in the blood pressure signal, and flags the beat as abnormal if its features,
which include beat duration, systolic, diastolic, mean, and pulse pressures, do not
fall within physiologically possible ranges. Zong et al. [33] had previously created
a blood pressure signal quality metric, wSQI, which uses fuzzy logic to asses the
extent to which the features of each ABP pulse fall within physiologically normal
ranges, yielding a continuous signal quality index value between 0 (abnormal) and 1
(normal). Li et al. [15] used the two ABP signal quality measures for robust heart
rate estimation from simultaneously recorded ECG and ABP waveforms by weighing
each beat’s jSQI value by wSQI .
Zong et al. [33] note that while jSQI and wSQI are successful at assessing signal
quality, they are limited by artifacts of the ABP measurement, such as those due to
catheter flush. The availability of the arterial blood pressure waveform poses further
limitations. We estimate that only 60% of adult ICU patients have ABP simultane-
ously recorded with ECG, due to the invasive nature of the measurement and simply
because not all patients require arterial blood pressure monitoring. Zong et al. [33]
suggest that the false critical ECG alarm suppression rate can be improved if the
ECG is compared to multiple cardiac function indicators, so that if one signal is of
poor quality, an alternate signal can be consulted. One source for such information
is the photoplethysmogram (PPG) waveform, which is pulsatile and non-invasively
obtained from a pulse oximeter affixed to the patient’s finger (in the case of trans-
mission pulse oximetry) or adhered to the skin (in reflective pulse oximetry). The
PPG waveform has different noise characteristics from the arterial blood pressure
waveform due to the difference in measurement technique and sensor location. For
example, the PPG waveform measures blood flow further down the arterial tree from
the site of the ABP measurement; as a result, the waveform resembles a delayed and
16
low-pass filtered version of the ABP waveform. A situation where the PPG waveform
might provide more information than the ABP waveform to the alarm suppression
framework is illustrated in Fig. 1-1. Here, the electrocardiogram waveform exhibits a
premature ventricular beat pattern which triggered an alarm. The arterial blood pres-
sure waveform is too noisy to consult for information to verify the ECG alarm. The
photoplethysmogram waveform, however, shows low-amplitude beats in accordance
with the premature beats which are inefficient at pumping blood.
Just as use of the information in the ABP waveform required blood pressure signal
quality measures, incorporation of information extracted from the PPG waveform into
the ECG false alarm suppression framework of Clifford et al. [3] requires assessment
of PPG signal quality to avoid drawing misleading information from an artifactual
waveform. The algorithms introduced in this thesis form a signal quality assessment
scheme for the PPG waveform recorded by a pulse oximeter. These algorithms, which
perform artifact detection, pulse onset identification, and pulse feature extraction, can
be used to determine high-quality segments of the PPG waveform, which can be used
to imporove false ECG alarm suppression and reduce true alarm suppression.
1.2 The Photoplethysmogram Waveform
1.2.1 Pulse Oximetry
Since its invention in the 1970s and commercial development in the 1980s, pulse
oximetry has provided a non-invasive method of estimating functional oxygen satura-
tion of the blood in clinical settings. Oximetry is based on the fact that hemoglobin
absorbs light in limited frequency ranges.
Oxygen reversibly binds to hemoglobin in the blood in order to nourish tissues
in the peripheral regions of the body. The oxygen is released from the blood and
into the tissue at the capillary level of the cardiovascular system. When oxygen
reversibly binds to hemoglobin, the resulting shift in the distribution of electrons
in the hemoglobin molecule causes its optical properties to change [6]. In particu-
17
Figure 1-1: Using the photoplethysmogram to corroborate ECG alarms. In this
segment of simultaneously recorded electrocardiogram, arterial blood pressure, and
photoplethysmogram waveforms, an ECG monitor would issue an arrhythmia alarm
due to premature ventricular beats. In this case, we would not be able to corroborate
the alarm by consulting the arterial blood pressure waveform because of noise in
the channel. However, the photoplethysmogram waveform does not exhibit noise
and the morphology of its beats can be related to the shapes and timing of the
electrocardiogram QRS complexes.
lar, oxygenated hemoglobin (O2Hb) absorbs visible light in the blue region, making
blood appear red. Reduced or deoxygenated hemoglobin (RHb) absorbs light at
most frequencies in the visible spectrum, making blood appear dark (or blue when
viewed through the layers of the skin). Permanent binding of carbon monoxide to
hemoglobin, forming carboxyhemoglobin (COHb), and the binding of ferric ions to
hemoglobin, forming methemoglobin (MetHb), also cause the hemoglobin absorption
spectrum to shift for various frequencies of light. As illustrated in Fig. 1-2, the light
absorption of O2Hb and RHb differ most significantly in the red and near-infrared
regions [27]. Pulse oximetry devices typically study the absorption of at least two
18
wavelengths of light, at approximately 660nm and 940nm, by measuring the amount
of light transmitted through or reflected from perfused tissue such as that found in
the finger, earlobe, or on the forehead.
Figure 1-2: Absorption spectrum of hemoglobin species. Transmitted light absorp-
tion of oxyhemoglobin and deoxyhemoglobin (reduced hemoglobin) differs most sig-
nificantly in the red and near-infrared frequencies. Note the extinction coefficients
are plotted on a logarithmic scale. Adapted from Figure 2 in [27].
Transmission pulse oximetry is based on an estimation of the Beer-Lambert Law,
which states that the intensity of light trasmitted through a material is proportional
to the intensity of incident light and exponentially related to the amount of light
absorbed. The amount of light absorbed by a sample, A, is a dimensionless quantity
defined in terms of the light intensity in the presence of the sample, I(in W/m2),
and in the absence of the sample, I0, as
A=log( I
I0
),(1.1)
19
and is linearly proportional to the extinction coefficient and path length, expressed
as
A=cl, (1.2)
where is the molar absorptivity (in m2/mol) of the sample, cis the concentration (in
mol/m3) of the hemoglobin species, skin, muscle, and bone, and lis the path length
in meters of the transmitted light [6].
In transmission pulse oximeters, one light-emmitting diode of each wavelength,
660 nm and 940 nm, shines incident at roughly 90to the outer tissue (typically
through the nail on the back of a finger tip for adults), and the intensity of the
transmitted light is detected on the opposite side (typically the finger pad). For each
frequency of incident light, the absorption can be expressed as a sum of absorption
due to O2Hb, RHb, COHb, and MetHb, as well as absorbtion by other non-blood
sources, such as surrounding tissues. Finger probe pulse oximeters operate under the
assumption that the path length lmaintains a steady “direct current” (DC) value due
to venous and arterial blood, as well as an alternating “current” (AC) component due
to the expansion of the capillaries as each wave of blood is pumped from the heart
and flows through the vasculature. The resulting absorption waveform is illustrated
in Fig. 1-3. It has been noted that the AC component of these absorption waveforms
account for less than 1% of the total light absorbed by the perfused tissue. Absorption
measurements are highly susceptable to any change in the material surrounding the
pulsating arterial vasculature, including the disturbance of muscle, skin, and venous
blood in response to motion [22].
The pulsatile PPG waveform displayed on ICU monitors is a dimentionless quan-
tity computed from a ratio comparing the AC amplitude to DC light absorption of
the red and infrared wavelengths as follows [6, 27]:
20
Figure 1-3: Light absorption waveform in inhomogenious tissue. In pulse oximetry,
the AC component is due to the varying path length when the arterial vasculature
expands during a pulse. Note this AC component only accounts for approximately
1% of the total absorption. Adapted from Figure 3 in [27].
R=dA660nm/dt
dA940nm/dt =AC660nm/DC660nm
AC940nm/DC940nm
.(1.3)
Ris a pulsatile waveform taking values between 0 and 1, similar in appearance to the
ABP.
From this ratio, an estimate of functional oxygen saturation in arterial blood can
be made [6]. Substituting equations 1.1 and 1.2 into equation 1.3, and recalling that
the pulsatile component of the signal is due to movement of oxyhemoglobin, Rcan
also be expressed as follows:
R=o,660nmco+r,660nmcr
o,940nmco+r,940nmcr
(1.4)
where oand coare the molar absorptivity and concentration of oxyhemoglobin,
and rand crare the molar absorptivity and concentration of reduced hemoglobin,
respectively. Functional oxygen concentration is defined by %SaO2=co/(co+cr).
This allows us to express the theoretical relationship between %SaO2as follows [6]:
21
%SaO2=r,660nm r,940nmR
(r,660nm o,660nm)(r,940nm o,940nm)R(1.5)
Due to limitations to the Beer-Lambert law caused by light scattering in tissue and
pulse oximeter device characteristics, the true relationship between pulsatile estimate
of functional oxygen saturation, %SpO2and Ris empirically determined by fitting
data from human volunteers to an equation of the form S = (abR)/(cdR) [6].
This relationship is illustrated in Figure 1-4. In addition to the waveform, R, the
%SpO2, is reported as a percentage every second by the pulse oximeter. Normal
oxygen saturation levels range between 90 and 95%.
Figure 1-4: Typical pulse oximeter calibration curve, illustrating the relationship
between the measured ratio of fractional changes in light intensity at two wavelenths,
R, and estimated oxygen saturation %SpO2. Adapted from Figure 4 in [27].
22
1.2.2 Waveform Morphology
The photoplethysmogram waveform is similar in shape to the arterial blood pressure
waveform, but has several morphological differences which prevent simple use of the
wSQI and jSQI algorithms on PPG. As noted earlier, the PPG and ABP waveforms
have different scales, and the amplitude of PPG waveform typically ranges from 0 to
1 rather than from 30 to 300 mmHg. There is no direct meaning for low or large
pulse amplitudes in the PPG waveform. The PPG amplitude can be modulated
by respiratory activity, as with the ABP. In processed waveforms, the amplitude is
somewhat arbitrary due to automatic gain controls by the electronic monitor. The
variability in pulse-to-pulse time follows the activity of the heart. When the signal
is of good quality, the PPG pulse ampliutude varies closely with the stroke volume
of the heart on a beat-by-beat basis, and with respiration (through respiratory sinus
arrhythmia) [17]. The onset of each PPG pulse follows the onset of the QRS complex
in the electrocardiogram and the onset of the corresponding pulse in radially-measured
ABP. This can be quantified by the pulse transit time (PTT), which is computed as
PTT =tPPO tecgQRS ,(1.6)
where tPPO is the PPG pulse onset time and tecgQRS is the onset time of the cor-
responding QRS complex in the electrocardiogram, which should occur between the
current and last PPG pulse [5].
1.2.3 Artifacts
There are several limitations to the accuracy of pulse oximetry, including attenuation
due to poor perfusion, skin pigmentation, and nail polish [24]. Inaccurate oxygen
saturation values in certain types of anemic patients are due to modified hemoglobin
which cannot be characterized by the normal hemoglobin light absorption spectrum.
Artifacts due to ambient gas or fluorescent lighting have also been of concern, espe-
cially for those oscillating frequencies which are near the harmonic frequencies of the
pulse oximeter’s LED pulsations [24]. The waveform is subject to arbitrary baseline
23
shifts and to sudden amplitude changes due to the monitor’s automatic gain control.
Noise in the signal may cause the amplitude of the PPG waveform to saturate at a
maximum or minimum value, or to rest at some random fixed value. However, the
artifacts of largest concern are caused by motion of the sensor relative to the skin
(generally due to patient movement) [12, 25, 22].
Researchers have investigated several methods for PPG artifact reduction, which
can be categorized into three types of approach: stationary filtering based on fre-
quency content [12], adaptive filtering based on energy changes in the waveform [4, 30],
and adaptive filtering based upon data from an external sensor [25, 7, 28, 29]. Hayes
et al. used spectral analysis to determine the motion artifact frequency range to be
greater than three times the PPG fundamental frequency (heart rate); signal quality
was then quantified by taking the proportion of artifact signal power to total signal
power [12]. Coetzee et al. used recursive-least-squares adaptive filtering of patient
waveforms with a synthetic reference signal to reduce noise and reconstruct waveforms
[4]. Once artifacts have been identified, Kalman filters can be used to extract autore-
gressive coefficients for interpolation and smoothing of noisy pulse wave segments
[30]. The performance of this method depends on the order of the autoregressive
model. Sokwoo et al. have characterized motion artifacts by designing a snug-fitting
ring sensor equipped with an accelerometer for fingerbase PPG measurements [25].
Adaptive filtering based on Laguerre models has been used to characterize the re-
lationship between the PPG waveform and acceleration of the hand and finger, and
to remove PPG motion artifacts [7, 28, 29]. Identification of PPG segments with
poor signal quality attributed to motion artifact has been achieved by comparing
the pulse rate obtained from the PPG waveform to the ECG-derived heart rate [22].
Gil et al. used Hjorth parameters to estimate the dominant frequency and spectral
bandwidth of PPG waveforms measured from pediatric patients while in sleep, and
applied thresholds to mark regions of gross artifact [8, 9]. However, few others have
used these artifact reduction techniques to identify artifact types and create signal
quality measures. None have studied PPG signal quality in the context of adult ICU
patients, or under conditions of arrhythmias. We believe the stationary and adaptive
24
filtering approaches can be combined for more robust artifact detection.
1.3 Overview of Thesis
The goal of this thesis is to improve performance of ICU bedside monitors by suppres-
sion of false critical ECG arrhythmia alarms through the use of information derrived
from simultaneously acquired PPG and ABP waveforms. Augmentation of the false
alarm suppression framework presented by Clifford et al. [1, 3] to employ the PPG
waveform requires both feature extraction and signal quality assessment. Two algo-
rithms have been created for this purpose. Chapter 2 introduces the pSQI algorithm,
which employs spectral analysis to detect large artifacts for PPG signal quality assess-
ment. Chapter 3 introduces the aP P G algorithm, which employs time-series analysis
to detect PPG pulse onsets. The results of these two algorithms are incorporated into
a new false alarm suppression framework, which is described in Chapter 4. Evaluation
of these methods and a discussion of improvements can be found in Chapter 5, as
well as a discussion of future research efforts.
25
THIS PAGE INTENTIONALLY LEFT BLANK
26
Chapter 2
Signal Quality Assessment
The use of the photoplethysmogram waveform for electrocardiogram alarm corrobo-
ration requires a guarantee of the absence of artifact in the PPG waveform. Signal
quality assessment is therefore a necessary component of the false alarm suppression
framework.
2.1 Previous Work
Our PPG signal quality assessment is based on the identification of artifact periods
using spectral power characteristics, as performed by Gil et al. [8, 9]. ornmo et al.
estimated the dominant frequency and half-bandwidth of the spectral distribution of
the waveform using Hjorth parameters [13, 14]. Gil et al. thresholded these parameter
values to identiify periods of major artifact.
The Hjorth parameters characterize a time signal in terms of its amplitude, time
scale, and complexity. The parameters of a discrete signal, x[n], where nis the sample
number, are derived from the moments of the power spectrum Sx(e ), where ωis
the frequency in radians [13, 14]. The ith-order spectral moment is defined as
¯ωi=Zπ
πωiSx(e ). (2.1)
Since the power spectrum is symmetric about the ω= 0 frequency axis, the odd
27
moments are all zero. However, the even moments can be used to estimate the shape
of the power spectrum of the signal.
We assume that the signal x[n] is a sampled version of a continuous time signal
xc(t) with sampling period Ts, such that x[n] = xc(nTs) for n= 0,1, . . . , N 1. The
spectral moments can be computed from the mean power of xc(t) and its derivatives,
¯ω0=Zπ
πSx(e )= 2πE[xc2(t)],(2.2)
¯ω2=Zπ
πω2Sx(e )= 2πTs2E
dxc(t)
dt !2
,(2.3)
¯ω4=Zπ
πω4Sx(e )= 2πTs4E
d2xc(t)
dt2!2
,(2.4)
where E[y] indicates the calculation of the expectation of the argument, y. Note that
the zeroth moment corresponds to the variance, σ2
a, of the amplitude of the zero-mean
signal x[n]. Similarly, the second moment corresponds to the variance, σ2
d, of the slope
values of the signal, and the fourth moment corresponds to the variance, σ2
dd, of the
rate of change of slope in the signal.
The first Hjorth parameter (termed activity),
H0= ¯ω0,(2.5)
gives a measure of mean signal power. The second Hjorth parameter (termed mobility)
is defined as
H1=s¯ω2
¯ω0
.(2.6)
From a time-domain perspective, H1gives a measure of the standard deviation of
the slope of x[n] relative to the standard deviation of the amplitude. As a power
ratio, this parameter becomes a measure of frequency variance of the power spectral
density. The third Hjorth parameter (termed complexity) is expressed as
28
H2=s¯ω4
¯ω2
¯ω2
¯ω0
.(2.7)
The first term in the difference can be interpreted as the mobility or frequency variance
of the power spectral density of the first derivative of x[n], where the signal power
has been redistributed to the higher frequencies. The complexity parameter therefore
represents the difference between the frequency variance of the first derivative and
the frequency variance of the original signal. In the time domain, this parameter can
be interpreted as the variance of the curvature values during one period with respect
to the variance of the slope values during that period. A rapidly varying signal with
complex morphology, such as high-frequency noise, will exhibit more variance in the
curvature of the signal than a smoothly-varying signal such as a sinusoid.
Hjorth parameters can be efficiently computed in the time domain because the
spectral moments can be computed from the first and second derivatives of the time
series [13]. For discrete signals, these derivates are approximated by the first and
second difference equations, such that
x(1)[n] = x[n]x[n1],(2.8)
x(2)[n] = x[n+ 1] 2x[n] + x[n1],(2.9)
where
dixc(t)
dtix(i)[n]
Tsi.(2.10)
The interpretation and use of the Hjorth parameters for spectral estimation can
be clarified through an example. Suppose the input signal is a pure sinusoid with
fundamental frequency 1 Hz. The power spectrum of this signal consists of two
impulses centered at the positive and negative fundamental frequency of the sinusoid.
The slopes of the input signal are taken from the first derivative of the signal, which
is the cosine function. The variance of the slope values generated over one period
29
with respect to the amplitude of the signal during that period is expressed by the
mobility parameter, which should equal 1 for this signal because the variance of
the sine and cosine values is the same. From a frequency perspective, the mobility
parameter describes the frequency of the signal. (The mobility will be greater than 1
for sinusoids of higher frequencies, and less than 1 for sinusoids of lower frequencies,
due to scaling of the signal amplitude when a derivative is taken.) If the the signal
is a pure sinusoid, the complexity parameter should equal zero, because the second
derivative of the signal carries the same fundamental frequency as the first derivative
and the signal itself. The mobility of the first derivative is also 1, and because
the complexity parameter describes the difference between the frequency variance of
the first derivative and the frequency variance of the original signal, the complexity
parameter equals zero.
Because the Hjorth parameters are based on the concept of variance, they also
exhibit the additive properties of variance [13]. For example, in the case where the
signal x[n] represents a superposition of sinusoids, the mobility parameter will provide
some weighted measurement of all the present fundamental frequencies. If the signal
is periodic but not purely sinusoidal, the power spectrum will exhibit a peak asso-
ciated with the main frequency of the signal but will also have non-zero bandwidth.
The mobility parameter, as the frequency variance of the power spectrum, will not
represent this main frequency, and the complexity parameter will have a non-zero
value, reflecting the extent to which the morphology of the signal deviates from that
of a sinusoid.
In the analysis of physiologic signals such as the PPG waveform, the input sig-
nal is periodic but near-sinusoidal. This means the power spectrum has a resonant
frequency related to the heart rate, and has some non-zero bandwidth. Gil et al.
[8, 9] used the mobility parameter, H1, to estimate the dominant frequency of the
signal, and used the complexity parameter, H2, to estimate the half-bandwidth of
the PPG power spectrum. The H1value does not provide an accurate estimate of
dominant frequency or heart rate, as is illustrated in Figure 2-1. However, waveforms
with physiologically normal morphology and heart rate should exhibit power spectral
30
densities with most of their mass within a certain range.
Gil, Vergara, and Laguna [9] describe a PPG artifact detector which employs
mobility and complexity of the PPG waveforms. At each sample, H1and H2were
determined from estimates of the moments of a P-sample window of PPG data:
ˆ
¯ωi[n]2π
P
n
X
k=n(P1)
(x(i/2)[k])2, i = 0,2,4 (2.11)
where x(i/2)[k] is the i/2 derivative of x[k].
Figure 2-1: Hjorth parameter calculations for PPG segments at various heart rates.
Printed with each power spectrum is the mean of the H1and H2Hjorth parameters,
which were calculated using non-overlapping 2s and 4s windows from the zeroth,
second, and fourth moments of the power spectrum of the PPG waveform according
to Equation 2.11.
In their study of PPG data from 26 children, Gil et al. [8, 9] found that the PPG
signal is of high quality when mobility (H1) lies within a range whose upper threshold
31
is specified by η1uand whose lower threshold is given by η1l. The signal is considered
to be of good quality when complexity (H2) lies below a threshold, η2. Conversely,
artifactual regions are portions of the signal where the dominant frequency differs too
much from the heart rate, H1< η1lor H1> η1u, or where the power spectrum is too
wide, H2> η2.
The thresholds used to detect artifact, η1l,η1u, and η2are set at a static value
relative to ˜
H1and ˜
H2, the medians of the H1and H2values, which are computed
over the entire length of the provided data:
η1l=α1+˜
H1,(2.12)
η1u=α2+˜
H1,(2.13)
η2=α3+˜
H2.(2.14)
In the work of Gil et al.,α1=1, α2= 1.4, and α3= 3.
A single recording of data provided to the algorithm could be minutes long, or
it could be hours long. This method for setting thresholds thus assumes that the
majority of the recording contains signal with no artifact, and the heart rate and
rhythm are normal and stable. If the heart rate varies significantly over the interval,
portions of the segment may get marked as artifact even if the signal is of high quality.
Furthermore, the thresholds have been chosen based on data from pediatric patients
while in sleep, where motion artifacts are less likely to appear. The threshold settings
for artifact determination therefore may not hold for adult ICU patients, which are
the focus of this study.
2.2 Modifications for Prototype Artifact Detector
We have implemented the artifact detection algorithm of Gil et al. [8, 9], which we will
refer to as pSQI , on PPG data from the MIMIC I [20] and MIMIC II [23] databases.
32
This prototype system was used to provide a rough estimate of the total amount
of good quality PPG data available for use, as well as to screen for good quality
PPG segments while testing the aP P G pulse onset detection algorithm, described in
Chapter 3.
We have created a binary signal quality index, SQI, that takes value 0 for arti-
factual segments, and takes value 1 for good-quality segments. An illustration of the
performance of the artifact detector is shown in Figure 2-2 on page 34. Note that for
the temporal window over which H1and H2are calculated, Gil et al use a window size
of 5 seconds, and we use non-overlapping windows of length 2 seconds. The choice of
2 seconds is to provide the shortest temporal window possible which would capture at
least one pulse at 30 beats per minute or faster. When the SQI value drops from 1 to
0, indicating artifact, it does so at the back (earliest) edge of the P-sample window.
The SQI steps from 0 to 1 at the leading (latest) edge of the window. In other words,
the output of the SQI detector takes a safe harbor approach and labels any section
that may contain some artifact as artifactual, even though there may be some good
quality data near the start and/or end of the window. Furthermore, if two artifactual
segments are not separated by more than 5 seconds of clean data, they are fused into
one longer artifactual period. Therefore regions with SQI equal to 1 are unlikely to
have artifactual data in them, even at the region’s edges.
The prototype system has three main limitations. First, the thresholding mecha-
nism is not adaptive, and assumes that the majority of a record contains clean data,
since the thresholds are set relative to the mean Hjorth parameter values across all the
segments in a record. Secondly, the thresholds were determined from data recorded
in a pediatric sleep study and should not be applied to adult ICU patient data with-
out further investigation. The age, activity, condition, and treatment of adult ICU
patients differs systematically from sleeping pediatric patients. Third and most im-
portantly, while regions of normal sinus rhythm and normal pulse morphology are
marked as having high PPG quality, periods of high PPG signal quality (i.e. clearly
discernable beats despite low amplitude or atypical morphology) recorded when the
patient was suffering from a cardiac arrhythmia are often marked as artifact. Specif-
33
Figure 2-2: PPG artifact detection based on Hjorth parameters, as described in [9] on
MIMIC II patient record a44545. Non-overlapping window size = 2 s. (a) Mobility,
H1[n] (see Eqn. 2.6), with thresholds η1land η1u, outside of which the signal is
considered artifactual. (b) Complexity, H2[n] (see Eqn. 2.7), with spectral width
threshold η2, above which the signal is considered artifactual. (c) PPG waveform
(solid line) and signal quality (dashed line). The PPG waveform exhibits good quality
when η1l< H1< η1uand H2< η2.
34
ically, the prototypical pSQI algorithm marks periods of PPG with low heart rates
as artifact, thereby ignoring high quality signals measured during periods of brady-
cardia. We address these limitations by examining the H1and H2parameter values
in adult ICU patient data in the case of normal sinus rhythm and under various
arrhythmia conditions.
2.3 Adaptive Assessment of Signal quality
Our goal is to suppress false arrhythmia alarms in the ECG signals of ICU beside
monitors by consulting information in simultaneously recorded PPG signals. To do
this, we require a trust metric for the PPG waveform. By employing the Hjorth
parameters to analyze the PPG waveform recorded before an alarm, we can determine
if the PPG is of high enough quality to provide a reliable estimate of heart rate.
Due to the limitations discussed in the previous section, it is important to assess
PPG signal quality in the context of the specific arrhythmia alarm type which has
been generated by the monitor. Thus we examine the H1and H2Hjorth parameter
values computed from the PPG waveform segments recorded just before true and
false arrhythmia alarms are issued. Hjorth parameter thresholds must be determined
for each arrhythmia alarm type, since spectral content differs based on heart rate.
2.3.1 Structure and Availability of Waveform Data
For this research, the ICU photoplethysmogram waveforms are taken from the MIMIC
II database [23]. Through this database we have access to approximately synchronous
waveforms with combinations of respiration, electrocardiogram, arterial blood pres-
sure, and photoplethysmogram waveforms, sampled at 125 Hz, recorded by Phillips
CMS bedside patient monitors (Phillips Medical Systems, Andover, MA). The data
is organized by patient records, which also contain annotations of all ECG-, ABP-,
and PPG-issued alarms.
Each patient record (which contains data from an individual ICU visit) may be
broken into several record segments of variable length. A new segment begins when-
35
ever the number or type of channels of data changes, the gain of any channel of data
changes, the waveform file becomes corrupt, the time stamps become non-contiguous
(due to network errors), or the data collection unit is stopped for a few minutes to
allow changing of disks. The types of waveforms in one segment of the record may
not necessarily be present in a different segment of the same record. A total of 20,931
segments of varying lengths from 2,997 patient records are accessible. Only 618 of
these records contain electrocardiograms which triggered critical (life-threatening)
ECG arrhythmia alarms, with 6,977 critical alarms total.
We have categorized each waveform segment based upon its signal content, and
created groups for segments containing ECG, ABP, PPG, and waveforms labeled
“unknown.” Inspection of these waveforms shows they are mostly mislabeled PPG
waveforms, so in estimating the amount of available photoplethysmogram data, we
take the union of the set of segments containing labeled PPG waveforms with the
set of segments containing a waveform labeled “unknown.” The number of hours
of available waveforms from these records for training and testing the signal quality
algorithm and ECG false alarm suppression algorithm is summarized in Table 2.1 and
illustrated in Figure 2-3. There are at least 47,581 hours of simultaneous PPG and
ECG data available for training and testing the signal quality algorithm and ECG
false alarm suppression algorithm. If 50,520 represents the total hours of all the
patient stays, then pulse oximeter data is available 94% of the time. (Note that there
are only 17,833 hours of simultaneous ECG and ABP waveforms, representing ABP
availability during only 35% of the patient record hours. This is much less than that
available from the PPG). However, when we add the criterion that life-threatening
arrhythmia alarms must be present at some point in the record, we reduce the number
of cases by 18% and available hours of waveform data by 35%. Approximately 11,231
hours of simultaneous ECG, ABP, PPG from 272 cases are available for training
and testing the PPG pulse onset detector and to evaluate performance of a PPG-
enhanced false alarm suppression framework. After requiring that the record is longer
than 5 minutes, contains critical arrhythmia (asystole, extreme bradycardia, extreme
tachycardia, ventricular tachycardia, or ventricular fibrillation/tachycardia) alarms,
36
Table 2.1: Estimated hours of available waveform data
Waveforms All Available Records Records with Alarms
No. Cases No. Hours (%) No. Cases No. Hours (%)
ECG 756 50,520 (100%) 618 32,897 (65%)
ECG & PPG 728 47,581 (94%) 596 31,325 (62%)
ECG & ABP 315 17,833 (35%) 283 11,885 (24%)
ECG & ABP & PPG 303 16,654 (33%) 272 11,231 (22%)
and is not the record of a patient with an intra-aortic balloon pump, the final number
of cases considered is 181.
Figure 2-3: Waveform data available with critical electrocardiogram alarms. Of the
618 cases with critical ECG alarms, 272 cases have simultaneously recorded ECG,
ABP, and PPG waveforms.
2.3.2 Preparation of Alarm Data
From the MIMIC II database, ICU patients have been selected whose records in-
clude simultaneously recorded ECG, ABP, and PPG waveforms and some number of
life-threatening cardiac arrhythmia alarms, namely asystole, extreme bradycardia, ex-
treme tachycardia, ventricular tachycardia, and ventricular fibrillation. Each of these
alarms were annotated independently as True, False, or Indeterminable by one signal
37
processing expert with over a decade of experience in analyzing such data and one
graduate student with graduate level training in cardiac electrophysiology [18, 11].
One physician with several decades of electrocardiographic interpretation experience
adjudicated the annotations. The annotations and adjudications were made by re-
viewing all ECG, ABP, and PPG waveforms surrounding each alarm onset over any
length of window size desired (but generally 30 seconds) using open-source waveform
viewing software (‘WAVE’, available at PhysioNet.org [19]). Patients with intra-aortic
balloon pumps were excluded from this study. The final “gold standard” annotated
alarm set included a total of 4,012 alarms from 181 ICU patients.
The alarm category of ventricular fibrillation/tachycardia yielded no true ven-
tricular fibrillation annotations, where the ECG waveform accompanying the alarm
exhibited uncoordinated ventricular contraction and the ABP and/or PPG showed
no pulsatile activity until the patient received a defibrillating shock. The alarm was
always triggered under circumstances of rapid ventricular tachycardia, which usually
degenerates into ventricular fibrillation. These ventricular fibrillation/tachycardia
alarms were therefore annotated in the same manner as ventricular tachycardia alarms
and were combined with alarms from the ventricular tachycardia category. That is, if
ventricular fibrillation was not present, but ventricular tachycardia was, the alarm was
marked as true. Similarly, if the associated waveforms demonstrated neither ventric-
ular fibrillation nor ventricular tachycardia, the ventricular fibrillation/tachycardia
alarm was annotated as false.
A separate signal quality study was conducted for each of the alarm types. Patients
exhibiting the alarm in question were ranked by the number of alarms in the record
and sorted into training and test sets. Each set had an equal number of patients
and roughly equal number of alarms. The number of patient records and the relative
frequency of true and false alarms of each type in the training and testing sets are
detailed in Tables 2.2, 2.3, and 2.4. Compared to the data set in the study performed
by Aboukhalil et al. [1], our data set has a higher percentage of ventricular tachycardia
alarms, and fewer extreme bradycardia and extreme tachycardia alarms. The false
alarm rates in our data set are similar, except for extreme tachycardia alarms, which
38
Table 2.2: Annotated critical ECG arrhythmia alarms in gold standard database. For
example, there are 29 true asystole alarms, indicating that 1.2% of all alarms in the
database are true asystole alarms, and that 7.8% of all asystole alarms in the data
set are true.
Alarm Type No. Patients
No. Alarms
All True False
(% All) (% All) (% Type) (% All) (% Type)
Asystole 95 639 50 589
(15.9) (1.2) (7.8) (14.7) (92.2)
Extreme 49 282 174 108
Bradycardia (7.0) (4.3) (61.7) (2.7) (38.3)
Extreme 66 832 762 70
Tachycardia (20.7) (19.0) (91.6) (1.7) (8.4)
Ventricular 146 2259 1194 1068
Tachycardia (56.3) (29.7) (52.7) (26.6) (47.3)
Total 181 4012 2177 1835
(Averages) (13.6) (53.5) (11.4) (46.6)
appear less frequently in our alarm collection.
2.3.3 Preparation of Normal Sinus Rhythm Data
To provide an understanding of the spectral distributions of PPG signals during sinus
rhythm, and to provide a set of data for training a pulse onset detection algorithm,
we identified a large amount of clean PPG data recorded during sinus rhythm.
From the MIMIC II database, 748 half-minute segments of non-artifactual PPG
signals exhibiting normal sinus rhythm were screened for high signal quality using the
prototype pSQI algorithm and examined by eye to ensure the lack of gross artifact,
signal dropout, and indications of arrhythmia. Beat onsets were detected (using a
pulse onset detection algorithm described in Chapter 3), and a vector of beat-by-beat
instantaneous heart rates was formed for each PPG segment. Those segments which
exhibited mean instantaneous heart rates between 60 and 85 beats per minute, with
a standard deviation less than 5 beats per minute, were retained to yield 264 half-
minute epochs from 43 patients. The H1and H2parameters for each non-overlapping
2 s window of these epochs were recorded.
39
Table 2.3: Annotated critical ECG arrhythmia alarms in Training Set. For example,
in the training set there are 29 true asystole alarms, indicating that 1.3% of all alarms
in the training set are true asystole alarms, and that 8.3% of all asystole alarms in
the training set are true.
Alarm Type No. Patients
No. Alarms
All True False
(% All) (% All) (% Type) (% All) (% Type)
Asystole 48 349 29 320
(15.3) (1.3) (8.3) (14.0) (91.7)
Extreme 25 205 131 74
Bradycardia (9.0) (5.7) (63.9) (3.2) (36.1)
Extreme 33 519 498 21
Tachycardia (22.7) (21.8) (95.9) (0.9) (4.1)
Ventricular 73 1213 711 502
Tachycardia (53.1) (31.1) (58.6) (22.0) (41.4)
Total 127 2286 1369 917
(Averages) (15.0) (56.7) (10.0) (43.3)
Table 2.4: Annotated critical ECG arrhythmia alarms in Test Set. For example, in
the test set there are 21 true asystole alarms, indicating that 1.2% of all alarms in
the test set are true asystole alarms, and that 7.2% of all asystole alarms in the test
set are true.
Alarm Type No. Patients
No. Alarms
All True False
(% All) (% All) (% Type) (% All) (% Type)
Asystole 47 290 21 269
(16.8) (1.2) (7.2) (15.6) (92.8)
Extreme 24 77 43 34
Bradycardia (4.5) (2.5) (55.8) (2.0) (44.2)
Extreme 33 313 264 49
Tachycardia (18.1) (15.3) (84.3) (2.8) (15.7)
Ventricular 73 1046 480 566
Tachycardia (60.6) (27.8) (45.9) (32.8) (54.1)
Total 126 1726 808 918
(Averages) (11.7) (48.3) (13.3) (51.7)
40
2.3.4 Hjorth Parameter Assessment By Alarm Type
For each alarm in the training set, thirty seconds of PPG data prior to the alarm
were extracted and analyzed to compute the H1and H2parameters over 2 s non-
overlapping windows. The Hjorth parameters were then sorted by associated alarm
type and condition (the veracity of the alarm, true or false). Distributions of these
H1and H2values are illustrated using box and whisker plots in Figures 2-4 and 2-5.
The mobility parameter (H1) estimates the dominant frequency of the PPG sig-
nal, which is noticeably lower in the case of true bradycardia alarms compared to
waveforms at normal sinus rhythm or faster heart rates. False bradycardia alarms
are accompanied by waveforms with the dominant frequency in the same range as
those exhibiting normal or fast heart rates. True asystole alarms are accompanied
by a wide range of dominant frequencies, indicating the high prevalence of noise and
gross artifact, or missing signal on the PPG channel, while PPG waveforms measured
during false asystole alarms exhibit a dominant frequency in the range of normal and
fast heart rates.
The complexity parameter (H2) shows more promise for distinguishing between
high signal quality PPG waveforms in true and false conditions of extreme tachycardia
or ventricular tachycardia alarms. For the ventricular tachycardia category, PPG
waveforms accompanying true alarms exhibit more “band-limited” power spectra than
those alongside false alarms. Once again, the wide range of complexity parameters
for true asystole alarms indicates wider bandwidth, which is associated with a flat,
DC-like signal.
A two-sample Kolmogorov-Smirnov test was performed between the true and false
distributions for each alarm condition and both Hjorth parameters. The results are
presented in Table 2.5. The H1distributions for extreme tachycardia alarms were
not significantly different. In every other case, the distributions were found to be
significantly different (p < 0.0001).
41
Figure 2-4: Box and whisker plot of mobility parameter (H1) distributions by alarm
type and condition (veracity). ASYS = asystole; BRAD = extreme bradycardia;
TACH = extreme tachycardia; VTAC = ventricular tachycardia.
2.3.5 Threshold Setting
As a first pass, we assume that most of the “mass” in the distributions of the H1and
H2parameters under true alarm conditions is calculated from clean PPG waveforms.
These segments most likely contributed to the annotation of the alarm as true. We
therefore use values of H1and H2which are at the edges of the distributions to
indicate poor signal quality.
In the remainder of this study, we examined 512 combinations of the Hjorth pa-
rameter thresholds, η1l,η1u, and η2. For each threshold we chose eight values spanning
the upper and lower interquartile ranges in the true alarm distributions. For η1l, we
chose eight uniformly spaced values between one and a half interquartile ranges below
the lower quartile and the median value of each true alarm H1distribution. For η1u,
42
Figure 2-5: Box and whisker plot of complexity parameter (H2) distributions by alarm
type and condition (veracity). ASYS = asystole; BRAD = extreme bradycardia;
TACH = extreme tachycardia; VTAC = ventricular tachycardia.
we chose eight uniformly spaced values between the median value and one and a half
interquartile ranges above the upper quartile of each true alarm H1distribution. For
η2, we chose eight uniformly spaced values between the median value and one and
a half interquartile ranges above the upper quartile of each true alarm H2distribu-
tion. The threshold ranges tested are summarized in Table 2.6, where each range is
distributed from the lower value to the upper value in eight equal increments. Each
of these 512 Hjorth parameter thresholds was used in the false alarm suppression
framework described in Chapter 4, and the thresholds which yielded the best false
alarm suppression rate on the training set was chosen for use in the algorithm.
43
Table 2.5: Results of Kolmogorov-Smirnov tests for H1and H2during true and false
alarms to be sampled from different distributions
Alarm Type H1significance H2significance
Asystole p < 0.0001 p < 0.0001
Extreme Bradycardia p < 0.0001 p < 0.0001
Extreme Tachycardia p= 0.01 p < 0.0001
Ventricular Tachycardia p < 0.0001 p < 0.0001
Table 2.6: Ranges of Hjorth parameter threshold settings tested for each alarm type
Alarm Type
Threshold Range
η1lrange η1urange η2range
Increment Increment Increment
Asystole [0 . . . 1.68] [1.68 . . . 4.96] [13.1. . . 20.6]
0.24 0.47 1.1
Extreme Bradycardia [0.493 . . . 1.35] [1.35 . . . 2.17] [2.39 . . . 5.79]
0.12 0.12 0.49
Extreme Tachycardia [0.443 . . . 1.87] [1.87 . . . 3.29] [3.83 . . . 8.99]
0.20 0.20 0.74
Ventricular Tachycardia [0.555 . . . 1.81] [1.81 . . . 3.08] [3.98 . . . 7.70]
0.18 0.18 0.53
2.4 Use of pSQI
The pSQI algorithm is used in a false alarm suppression framework to assess the
signal quality in the PPG waveform just preceding an ECG arrhythmia alarm. If
the signal quality is high, the PPG waveform exhibits spectral characteristics of the
cardiac arrhythmia in question, and we trust the heart rate estimated from that
segment. Additional logic can then be used to accept or suppress the issued alarm,
as in the works of Aboukhalil, Clifford, et al. [1, 3].
44
Chapter 3
PPG Pulse Onset Detection
In order to perform heart rate estimation and beat-by-beat extraction of PPG wave-
form features, the duration of each pulse must be determined. This can be achieved
by pulse onset detection, assuming the pulse lasts from one onset to the next, and no
beats are missed or erroneously detected.
3.1 Previous Work
For the simple purpose of heart rate estimation from the photoplethysmogram wave-
form, peak detection is a simple and effective method for pulse identification. Pulse
peak detection can be made robust to noise and movement artifacts if adequate filter-
ing and thresholding is applied, as demonstrated by Yu et. al [31]. However, studies
of irregular pulse morphology or rhythms require feature analysis on the whole pulse.
Detection of pulse onsets allows for pulse extraction and study of irregular pulse
morphology, as well as analysis of pulse transit time.
Zong et al. [32] have previously created the wABP algorithm to detect the onset
of arterial blood pressure pulses. Their algorithm passes the input blood pressure
waveform,xn, through a low-pass filter, then computes a slope sum function (SSF),
which enhances the upslope of each pulse in the waveform. The low-pass filter is a
second-order recursive filter with transfer function, frequency response, and difference
equation given by Equations 3.1, 3.2 and 3.3.
45
H(z) = (1 z5)2
(1 z1)2(3.1)
|H(ωT )|=sin2(3ωT )
sin2(ωT /2) (3.2)
yn= 2yn1yn2+xn2xn5+xn10 (3.3)
For each time point, i, the SSF, ziof the preceding w-sample window of the filtered
signal, yn, is computed as follows:
zi=
i
X
k=in
uk,uk=
yk,if ∆yk>0
0,if ∆yk0
(3.4)
where 1 + wiN, N is the total number of samples in the ABP waveform, and
yk=ykyk1. The SSF is then passed through a decision rule to determine the
occurrence of each pulse onset in the blood pressure waveform.
In the wABP algorithm, the decision rule has two components, which we will
refer to as the pulse initiation and pulse confirmation phases. In the pulse initiation
phase, the algorithm determines that a pulse is initiated if the SSF value is greater
than a threshold. The threshold is initialized by the average SSF value over the first 8
s of waveform data. To confirm that an ABP pulse with a strong upstroke is present,
the difference between the maximum and minimum values of the SSF in a 150 ms
window must exceed a static confirmation threshold. If both of these conditions are
met, the pulse onset time is noted, and further detections are prohibited during the
following 0.25 s refractory period. Adaptation of the threshold is achieved by lowering
the initiation threshold by a constant value if 2.5 s have passed without any initiated
detections, and by updating the initiation threshold according to the maximum SSF
value of each detected pulse. The performance of the wABP algorithm is illustrated
in Fig. 3-1.
In Zong’s original work [32], the wABP algorithm was not evalutated in the
presence of artifact or noise, and its performance was not evaluated during periods of
arrhythmia, where the morphology of the blood pressure pulses deviates significantly
46
Figure 3-1: Use of the Slope Sum Function to detect pulse onsets in the arterial blood
pressure waveform. Adapted from Figure 4 in [32].
from the morphology at normal sinus rhythm, and the algorithm is therefore expected
to present unusual behavior. The wABP algorithm contains no adaptations for heart
rate variability. Furthermore, the limited adaptivity of the SSF pulse initialization
threshold and the use of a static threshold on the range of the slope sum function
indicates that the algorithm expects pulses of a certain amplitude and assumes there
will be only small deviations in pulse pressure.
3.2 aP P G: Photoplethysmogram Pulse Onset De-
tection
The similarities between ABP and PPG pulse morphology prompted us to adapt the
wABP pulse onset detector for use on the photoplethysmogram waveform, which we
will refer to as aP P G. As in the original wABP algorithm, we have maintained a
window size of 128 ms (n= 16 samples for a signal sampled at 125 Hz) for computing
the SSF, which corresponds to the typical upslope duration of a PPG pulse under
normal sinus rhythm heart rates. We have scaled and offset the PPG waveform input
to resemble physiologic range for blood pressure measurements (in mmHg) in order
to take advantage of the existing low-pass filter in wAB P .
47
The amplitude of the PPG waveform can change for several reasons, including
vasoconstriction, variation in pulse volume due to arrhythmia conditions, or as ar-
tifacts of automatic gain changes in the bedside monitors. Further modifications
have therefore been made to allow aP P G to perform robustly in the presence rapid
amplitude changes.
The refractory period is set to 0.25 s by default, but is modified if provided with
an estimate of the prior heart rate estimate, by assuming the length of the refractory
period is 40% of the total pulse duration. For simplicity, we define pulse duration to be
the pulse-to-pulse interval. This modification anticipates longer inter-pulse intervals
during periods of true bradycardia or shorter inter-pulse intervals during periods of
extreme tachycardia, by lengthening or reducing the refractory period (respectively)
in the presence of a prior heart rate estimate.
To make the pulse detection algorithm robust to sudden gain changes in the PPG
waveform, both the SSF pulse initiation and the SSF pulse confirmation phases of the
decision rule have been made adaptive. As in wAB P , the pulse initiation threshold
is intialized to three times the average SSF value over the first 8 s of data. At each
confirmed pulse onset detection, the threshold is adapted according to a fraction, Tc
of the difference between the local maximum and minimum values of the SSF.
An appropriate change in slope in each PPG waveform pulse is determined using
the magnitude of the corresponding SSF pulse. This SSF pulse confirmation threshold
was set to a static value in w ABP . However, a sudden decrease of the PPG signal
gain proportionally decreases the slope of the PPG pulse onsets. The corresponding
slope sum function peaks will also be diminished in magnitude. The static threshold
developed for periods of higher PPG signal gain will miss these pulses of diminished
amplitude. Therefore we adapt the pulse confirmation threshold. Adapting by the full
amplitude of the SSF waveform results in missed detections when there is a sudden
decrease in gain, so for a more robust system we adapt by Tc= 70% of the most
recentltly detected SSF peak.
Two timer algorithms have been introduced, to adapt the respective thresholds in-
dependently. The first timer, also used in wAB P , lowers the pulse initiation threshold
48
by a constant amount if no pulse has been initiated for more than 2.5 s. The second
timer, new to aP P G, continuously decreases the pulse confirmation threshold by a
fraction, dc= 0.1% per time step (or 12.5% per second at 125 Hz), of the PPG wave-
form amplitude over the previous 5 s if no new pulse has been detected for more than
four refractory periods. To avoid false pulse detections due to low-amplitude artifacts,
a noise floor is set at half of the smallest expected true pulse amplitude. If the full
possible range for the PPG waveform is 0 to 1, we expect the lowest pulse amplitude
to be no smaller than 0.025, so we set the noise floor at 0.0125. The confirmation
threshold is reset based on half of the PPG amplitude over the previous 5 s if the
noise floor is reached.
The performance of the aP P G algorithm is illustrated in Figures 3-2 and 3-3.
3.3 aP P G Performance
To evaluate the performance of the prototype algorithm, we compared pulse onsets
detected in the PPG waveform using aP P G to a set of “chrome standard” beat
annotations from the ECG and arterial blood pressure waveforms. That is, we chose
locations where the ECG and ABP pulse detections agreed, and assumed a PPG
pulse should also be present within a set period of time.
3.3.1 Data Acquisition, Pre-processing, and Evaluation Setup
Thirty one patient records of variable length containing simultaneously recorded ECG,
ABP, and PPG signals are available the MIMIC I database [20]. Waveforms were
extracted from all patient records, yielding 1,099.85 hours of data total. The PPG
waveforms were pre-processed to note the start and end points of any instance of
flat-line artifact or signal dropout. The PPG waveforms were also screened for severe
motion artifacts using a prototype pSQI signal quality assessment scheme, described
in Chapter 2 of this thesis. Any beats or blood pressure pulse onsets detected in
segments of the record where the PPG had dropped out, was flat, or contained severe
artifact (such as maximum or minimum saturations or high-frequency noise), were
49
Figure 3-2: PPG pulse onset detection by aP P G under conditions of normal sinus
rhythm, asystole, and bradycardia. Under normal sinus rhythm, the amplitude of the
PPG waveform stays constant, the pulse confirmation threshold rests at 70% of the
pulse amplitude, and the aP P G algorithm detects all the pulses. In the examples of
asystole and bradycardia, the pulse confirmation threshold is decreased in expectation
of low-amplitude pulses. After 4 refractory periods (1 s) following the last pulse
detection, the pulse confirmation threshold decreases at a rate of dc·Fs= 12.5% per
second until the next pulse is detected and the confirmation threshold is adjusted to
the recent amplitude of the waveform. A noise floor is set at 0.0125 to avoid false
pulse detections.
50
Figure 3-3: PPG pulse onset detection by aP P G under conditions of tachycardia and
ventricular tachycardia. At the onset of sustained extreme tachycardia or ventricular
tachycardia (illustrated in the top and bottom traces, respectively), a sudden decrease
in pulse amplitude is observed. In the case of unsustained ventricular tachycardia
(illustrated in the center trace), intermittent rapid beats yield low-amplitude pulses
in the PPG. All three cases cause missed detections by the pulse onset detector. After
4 refractory periods (1 s) following the last pulse detection, the pulse confirmation
threshold decreases at a rate of dc·Fs= 12.5% per second until the next pulse is
detected and the confirmation threshold is adjusted to the recent amplitude of the
waveform.
51
excluded from this study. The remaining segments were presumed to contain clean,
good quality PPG data, and the corresponding pulse onsets were included in the
study.
Annotations for QRS detections in the ECG and pulse onsets in the ABP wave-
forms are both available for records in the MIMIC I database. A standard open-source
beat comparison algorithm, bxb [10], with an 800 ms match window to account for
pulse transit time was applied to these annotations to find beats appearing in either
the ECG or ABP waveforms. With matching beats counted only once, 6,058,072
beats were found in the two waveform types. ECG and ABP beats appearing while
the PPG waveform exhibited signal dropout, flat-line or other gross artifact (such as
maximum or minimum saturations or high-frequency noise) were excluded from the
analysis. The remaining 4,227,904 beats (detected from 1,033.4 hours of waveform
data) were used as a surrogate for a gold-standard reference set for evaluating the
performance of aP P G.
The aP P G algorithm was applied to the PPG waveforms, and detected 3,859,567
beats in the good-quality waveform segments. These beats were compared to the
surrogate reference set using bxb with an 800 ms match window.
3.3.2 Results
Of the beats detected in the ECG and ABP waveforms, 87.53% were also detected
in the PPG waveform by the algorithm. Of the beats detected by aP P G, 95.88%
were also annotated pulses in the ECG or ABP waveforms. The sensitivity and
positive predictive value of the aP P G algorithm performance over all 31 patients are
summarized in Table 3.1.
3.3.3 Discussion of Limitations
Of the pulses detected by the aP P G algorithm, 4.12% did not appear in the ABP
or ECG waveforms. However, 12.47% of the pulses which travel from the heart to
the periphery are not detected in the PPG. One explanation is that these pulses do
52
Table 3.1: Performance of aP P G on MIMIC I database. TP = true positive (PPG
detection also appeared in ECG or ABP); ; FP = false positive (PPG detection did
not appear in ECG or ABP); FN = false negative (Pulse detection in ECG or ABP
did not appear in PPG.
Record Length of TP FP FN Sensitivity PPV
Number Record (h) (%) (%)
466 68.7 249654 6678 26610 90.37 97.39
427 58.5 98134 2234 13323 88.05 97.77
444 54.0 290548 1023 6456 97.83 99.65
430 52.0 190945 815 15805 92.36 99.57
213 51.7 76198 463 7756 90.76 99.40
408 48.3 235952 2738 6931 97.15 98.85
224 46.9 237039 1288 4579 98.10 99.46
439 46.4 235428 2507 9350 96.18 98.95
411 45.6 115489 1238 19111 85.80 98.94
409 43.3 242864 3716 33991 87.72 98.49
454 42.8 120114 7137 41001 74.55 94.39
231 42.7 8537 30427 2836 75.06 21.91
449 42.5 157280 1207 10933 93.50 99.24
254 42.5 146822 1225 15282 90.57 99.17
484 42.0 183598 5444 28372 86.62 97.12
474 38.5 140112 18088 11089 92.67 88.57
442 35.1 97630 1864 46042 67.95 98.13
452 33.7 156110 4397 16679 90.35 97.26
451 31.2 75754 425 27078 73.67 99.44
477 30.0 76136 2911 48336 61.17 96.32
446 27.9 30607 35613 17739 63.31 46.22
216 27.2 106099 20146 13541 88.68 84.04
218 26.0 69079 1371 18099 79.24 98.05
414 25.1 48516 2029 30420 61.46 95.99
410 23.5 92253 1750 7868 92.14 98.14
211 21.6 51243 640 10191 83.41 98.77
230 19.0 47660 331 2370 95.26 99.31
041 14.3 59673 220 1721 97.20 99.63
417 12.2 17033 432 21189 44.56 97.53
472 8.7 40762 673 11561 77.90 98.38
220 1.2 3262 6 1114 74.54 99.82
Sum 1,033.4 3,700,531 159,036 527,373
Gross 87.53 95.88
Average 83.49 93.42
53
not reach the periphery with significant volume to be detected. The topology of
the cardiovascular system is such that if the heart beats with low cardiac output,
pulse volume and velocity is not sufficient to appear at the end of the arterial tree
where the PPG is measured. This cannot be concluded definitively by the analysis
presented here because the two reference beat sources are recorded using independent
sources and are subject to their own artifacts, and the signal quality metric assessment
system is not optimal. However, an estimate of ABP pulse pressure prior to each pulse
detection could be used to indicate if there is sufficient pulse volume.
The performance analysis could be made more accurate by improving the reference
database against which we compare the detected PPG pulse onsets. A large database
of expert annotated PPG pulse onsets over a wide range of physiological conditions
would be ideal, but creation of such a database is time consuming. A better surrogate
for a gold standard reference would be to choose beats appearing in both the ECG
and ABP waveforms, rather than the union of the two pulse detection sets used in
this study. The reference beat set could be further improved by ensuring the signal
quality of the ECG or ABP waveforms by using algorithms such as wSQI [33].
The PPG pulse onset detection algorithm performs moderately “well” compared
to wABP under conditions of normal and slow steady heart rates, as illustrated in
Figure 3-2. The common incidence of false pulse onset detections during periods of
Asystole can be avoided by turning off the pulse detecting feature if the range of the
PPG signal is not above a certain threshold, thereby ignoring both truly flat and
nearly flat PPG waveform segments.
In the current implementation, the adaptive threshold mechanism does not adapt
quickly enough to detect low-volume pulses which result from premature ventricular
contractions. Similarly, the algorithm takes too much time to adapt to low pulse
volume during sustained tachycardia or ventricular tachycardia. These limitations
are illustrated in Figure 3-3. If tachycardia is not sustained, the pulses are likely
to remain undetected. One improvement could be to use the electrocardiogram’s
QRS detections to adapt the pulse threshold. The algorithm also does not adapt
quickly enough to a common PPG artifact in which signal saturation to minimum
54
or maximum for less than one second triggers the monitor to reduce the gain of the
waveform by half or more for the next three to five beats.
The performance of aP P G therefore is promising for heart rate estimation. How-
ever the limitations posed by sudden gain changes and pulse volume variability could
be improved by further tuning or reference to the ECG. Next, we propose a method
for tuning the algorithm.
3.3.4 Future Work: Parameter Optimization and Testing
The PPG waveform pulse onset detector has two thresholds, the pulse initiation
threshold and pulse confirmation thresholds, which are adapted at separate rates
in order to obtain robust performance in the presence of arrhythmias. The pulse
confirmation threshold is updated by Tc, the fraction of the difference between the
local maximum and minimum values of the SSF. The pulse confirmation threshold
is continuously decreased by a fraction, dc= 0.1%, of the PPG waveform amplitude
over the previous 5 s if no new pulse has been detected for more than four refractory
periods. To determine the best adaptivity rate settings, Tcand dc, we should evaluate
the performance of the algorithm under various threshold combinations. The PPG
pulse morphology varies little when the heart rate and rhythm are constant but
deviates during periods of arrhythmia as the stroke volume changes. The aP P G
algorithm performs differently at different extremes of heart rate, so we should study
the threshold combinations during periods of steady heart rate, variable heart rate,and
a combination of steady and varying heart rate (to simulate real-life conditions).
To avoid over-training our data, data from patients with simultaneously recorded
ABP and PPG should be split into training and testing sets. A limited number
of epochs should be extracted from each patient record to ensure diversity of the
training and testing set. The epochs should be selected based on three criteria: (1)
that simultaneously recorded PPG and ABP waveforms are available, (2) that at least
90% of the blood pressure waveforms are marked as high signal quality by wSQI [33],
and (3) that the PPG segments contain no gross artifacts, signal dropout, or missing
data, as determined by the pSQI algorithm presented in Chapter 2.
55
The PPG segments should be sorted into two groups, one containing data at
normal sinus rhythm, and one containing data recorded while the patient has a cardiac
arrhythmia. Any PPG segment with a regular heart rate and clear pulses should be
sorted into the first category. Any segment with ectopic beats or high variation in
heart rate should be sorted into the second category.
The aP P G algorithm should be run over waveforms in both the steady heart
rate and arrhythmia categories with varying combinations of Tcand dc, and the
pulse onsets should be recorded. To create a reference set of annotations, wSQI [33]
should be run over the simultaneously recorded blood pressure waveform segments.
As before, a beat-by-beat comparison should be made between the recorded PPG
pulse onsets and the recorded blood pressure pulse onsets. Performance of aP P G
should be evaluated under each combination of parameters, and the Tcand dcvalues
which yield the best performance in each category should be applied to the algorithm
and run on the test sets. The performance observed will be the best possible scenario
of pulse onset detection, since the waveforms should not contain major artifacts.
Ideally, the steady and variable heart rate data should be mixed in a realistic
manner, and the aP P G algorithm should be tested on this third data set. Note that
this procedure requires an accurate assessment of the proportion of time which ICU
patients have steady versus variable heart rates.
An alternative evaluation procedure might take a stress-test approach, to study
the ability of aP P G to reliably detect pulses under varying levels of artifact. In this
procedure, one might take clean PPG and ABP data at various steady heart rates,
and inject controlled amounts of artifact to the waveforms. The artificial PPG arti-
facts could be generated in a manner similar to the artificial blood pressure artifacts
developed by Li et al. [16].
3.4 Use of aP P G
Once pulse onset times are established, the waveform can be broken down into indi-
vidual pulses and the morphology of each pulse can be considered. In this study, we
56
use pulse onset detections to estimate heart rates in PPG segments recorded while
an alarm is triggered in the electrocardiogram monitor.
57
THIS PAGE INTENTIONALLY LEFT BLANK
58
Chapter 4
A New False ECG Alarm
Suppression Framework Using the
PPG Waveform
Using the aP P G pulse onset detector and pSQI signal quality algorithms, we form
a new framework to suppress false critical ECG alarms by estimating heart rate and
studying pulse morphology in the PPG waveform.
4.1 Algorithm Architecture
In Section 1.1 and Appendix A, we reviewed the work of Aboukhalil et al., who
created a false ECG alarm suppression framework which utilises the arterial blood
pressure waveform to estimate heart rate and suppress heart-rate related alarms. This
algorithm first checks the signal quality of all detected beats in the ABP segment,
accepting the alarm if the signal quality is poor. Then the algorithm employs a system
of logic which considers the quality and morphology of each detected pulse, and the
instantaneous heart rate estimated from those pulses, to determine if the ECG alarm
should be accepted or suppressed. The details of the logic employed by this framework
are included in Appendix A, as well as the performance of the algorithm on the alarm
data described in Section 2.3.2.
59
We take a similar approach to assessing the validity of critical ECG arrhythmia
alarms using the PPG waveform. We utilise the windowing parameters which were
found to yield optimal alarm suppression performance in the study by Aboukhalil
et al. At the onset of each alarm, we extract a 17-second segment of PPG data,
extending from 13 seconds before the alarm to 4 seconds after the alarm. Next, we
apply the pSQI signal quality metric. In our framework we consider each type of
alarm separately, applying Hjorth parameter thresholds to determine signal quality
depending on the alarm type. If at least 80% of the windowed PPG signal is marked as
high quality according to the applied thresholds, then the heart rate is estimated from
the beat onsets detected by the aP P G algorithm. The ECG alarm is then flagged as
true or false depending on whether the heart rate is within the error tolerances found
to be optimal by Aboukhalil et al [1].
Because our PPG signal quality metric does not function on a beat-by-beat basis,
we do not study the number of beats with abnormal pulse morphologies. However, we
can use certain aspects of PPG phenomenology to support the logic based on heart
rate estimation. Specifically, we use the observation that during episodes of extreme
tachycardia and ventricular tachycardia, the pulse volume may decrease due to low
stroke volume. The sudden decrease in pulse amplitude causes several missed pulse
detections when the aP P G algorithm is used to mark the onset of each PPG pulse
(see Section 3.3.3), yielding long intervals between detected beats. Therefore, in place
of pulse morphology analysis, we construct a framework based on PPG-based heart
rate estimation and pulse duration.
At the onset of each critical ECG arrhythmia alarm, a 17-second PPG waveform
segment is extracted from 13 seconds prior the alarm to 4 seconds after. Alarms
where the PPG waveform is not available for this 17-second window were excluded
from the study. Next, the pSQI algorithm is used to assess the signal quality of the
extracted PPG segment. If less than 80% of the duration of the PPG waveform is
marked as high signal quality, the PPG signal quality of the 17-second segment is
judged to be poor and the ECG alarm is accepted as true. Otherwise, the aP P G
algorithm is applied to detect PPG pulse onsets. To reduce false pulse detections
60
by the aP P G algorithm, the noise floor on the slope sum function is increased from
Tn= 0.0125 to Tn= 0.02. The new value of Tnwas chosen to eliminate true asystole
alarm suppressions in the training set without decreasing false alarm suppression.
Finally, the following logic is employed to assess the validity of the alarm depending
on the alarm type. Figure 4-1 illustrates the algorithm architecture where the PPG
signal quality is sufficiently high. The windowing and thresholding parameters used
in the PPG-based false alarm suppression algorithm are summarized in Table 4.1.
Figure 4-1: False ECG Alarm Suppression Using the PPG Waveform. If at least 80%
of the 17-second PPG segment is high signal quality, this logic is used to determine
if the alarm should be accepted or suppressed. IPI = inter-pulse interval.
4.1.1 Asystole Processing
From the detected pulse onsets, the largest pulse-to-pulse interval (for the case where
the asystole resolves itself within the window) and the interval between the last de-
tected pulse and the end of the window (for the case where asystole lasted beyond
the analysis period) are calculated. If either of these intervals exceeds TA= 3 s, the
absence of beats is noted and the asystole alarm is accepted; otherwise it is suppressed.
61
4.1.2 Extreme Bradycardia Processing
Bradycardia alarms in the ECG waveforms are marked relative to a heart rate thresh-
old set on the monitor. The NB= 3 longest pulse-to-pulse intervals are calculated.
If the mean heart rate calculated from these intervals is within EB= 7 beats per
minute of the monitor’s heart rate, then the bradycardia alarm is accepted; otherwise
it is suppressed.
4.1.3 Extreme Tachycardia Processing
Tachycardia alarms in the ECG waveforms are marked relative to a heart rate thresh-
old set on the monitor. The (NT= 1) shortest pulse-to-pulse interval is calculated. If
the heart rate calculated from this interval is less than ET= 20 beats per minute over
the threshold set on the monitor, and the longest pulse-to-pulse interval is less than
TN oP ulse = 1.8 s long (indicating no missed detections due to loss in pulse volume),
then the tachycardia alarm is suppressed.
4.1.4 Ventricular Tachycardia Processing
A ventricular tachycardia alarm is accepted if either the longest pulse-to-pulse interval
is greater than TN oP ulse = 1.8 s (indicating missed detections due to loss in pulse
volume) or the heart rate calculated from the (NV T = 1) shortest pulse-to-pulse
interval exceeds RV T = 80 beats per minute. Otherwise, the alarm is suppressed.
4.1.5 Ventricular Fibrillation Processing
Ventricular fibrillation alarms are suppressed if the longest pulse-to-pulse interval is
less than TN oP ulse = 1.8 s (indicating no missed detections due to loss in pulse volume)
and the heart rate calculated from the (NV F = 1) shortest pulse-to-pulse interval is
less than RV F = 150 beats per minute. Otherwise, the alarm is accepted.
62
Table 4.1: Windowing and thresholding parameters in merged PPG-based false alarm
suppression algorithm. “PPI” signifies pulse-to-pulse interval. “N/A” signifies that
the parameter is not applicable.
Parameter Max. PPI HR error Intervals HR
length (s) margin for HR Threshold
Alarm Type (bpm) calc. (bpm)
Asystole TA= 3 N/A N/A N/A
Extreme N/A EB= 7 NB= 3 N/A
Bradycardia
Extreme TN oP ulse = 1.8ET= 20 NT= 1 N/A
Tachycardia
Ventricular TN oP ulse = 1.8 N/A NV T = 1 RV T = 80
Tachycardia
Ventricular TN oP ulse = 1.8 N/A NV F = 1 RV F = 150
Fibrillation
4.2 Optimization of Signal Quality Thresholds
To determine the optimal settings to use in the pSQI algorithm for each arrhythmia
type, the number of true and false alarms in the training set suppressed by this
PPG-based framework were recorded using 512 combinations of Hjorth parameter
thresholds, ηl
1,ηl
1, and η2, chosen from the distributions computed in Section 2.3.5 (see
Figures 2-4 and 2-5). Table 2.6 lists the thresholds settings which were examined.
The false and true alarm suppression rates are illustrated in Figures 4-2 through 4-5.
We chose parameter thresholds which maximized false alarm suppression while
minimizing true alarm suppression. This is possible for asystole, bradycardia, and
extreme tachycardia alarms, even if the set of thresholds where these two critera are
met is small (as is the case for extreme bradycardia and extreme tachycardia alarms).
However, for ventricular tachycardia alarms there was no combination of η1l,η1u, and
η2which maximized false alarm suppression while minimizing true alarm suppression.
This meant we could not suppress any number of false alarms without also suppressing
some true alarms in this category. For asystole, bradycardia, and extreme tachycardia
alarms, we choose to set the Hjorth parameter thresholds in the centroid of the volume
defined by those which yielded the highest false alarm suppression rate and the lowest
true alarm suppression rate. These volumes are marked using solid pink dots and
63
Figure 4-2: Effect of ηl
1,ηu
1, and η2on true and false alarm suppression rates during
asystole. In the top plot, solid pink dots mark the highest false alarm suppression
rate. In the bottom plot, solid blue dots mark the lowest false alarm suppression rate.
The set of thresholds which maximize false alarm suppression while minimizing true
alarm suppression are outlined.
solid blue dots, respectively, in Figures 4-2 through 4-5. For ventricular tachycardia
alarms, we choose to set the Hjorth parameter thresholds in the centroid of the volume
defined by those which yielded the highest false alarm suppression rate. The selected
threshold settings are summarized in Table 4.2.
64
Figure 4-3: Effect of ηl
1,ηu
1, and η2on true and false alarm suppression rates during
extreme bradycardia. In the top plot, solid pink dots mark the highest false alarm
suppression rate. In the bottom plot, solid blue dots mark the lowest false alarm
suppression rate. The set of thresholds which maximize false alarm suppression while
minimizing true alarm suppression are outlined.
Table 4.2: Optimal assignment of Hjorth parameter thresholds by alarm type using
training data
Alarm Type Threshold Setting
ηl
1ηu
1η2
Asystole 0.24 4.49 17.38
Extreme Bradycardia 0.92 2.17 5.06
Extreme Tachycardia 1.66 2.68 7.15
Ventricular Tachycardia 0.82 2.45 3.98
65
Figure 4-4: Effect of ηl
1,ηu
1, and η2on true and false alarm suppression rates during
extreme tachycardia. In the top plot, solid pink dots mark the highest false alarm
suppression rate. In the bottom plot, solid blue dots mark the lowest false alarm
suppression rate. The set of thresholds which maximize false alarm suppression while
minimizing true alarm suppression are outlined.
66
Figure 4-5: Effect of ηl
1,ηu
1, and η2on true and false alarm suppression rates during
ventricular tachycardia. In the top plot, solid pink dots mark the highest false alarm
suppression rate. In the bottom plot, solid blue dots mark the lowest false alarm
suppression rate. The set of thresholds which maximize false alarm suppression while
minimizing true alarm suppression are outlined.
67
Table 4.3: Performance of PPG-based false alarm suppression algorithm on new
MIMIC II data. Note that ventricular fibrillation/tachycardia alarms are combined
with and annotated as ventricular tachycardia alarms. “FA” signifies false alarm.
“TA” signifies true alarm. “FA Rate Before” and “FA Rate After” refer to the false
alarm rate before and after the modified suppression algorithm was used, respectively.
Alarm Type Data Set TA FA FA Rate FA Rate
Suppression Suppression Before After
Asystole
Training 0% 68.3% 91.7% 29.0%
Testing 9.5% 68.0% 92.8% 29.7%
Combined 4.0% 68.2% 92.2% 29.3%
Ext. Brad.
Training 0% 14.9% 36.1% 30.7%
Testing 0% 35.7% 18.0% 11.5%
Combined 0% 20.6% 28.3% 22.4%
Ext. Tach.
Training 0% 14.3% 4.1% 3.5%
Testing 2.3% 2.0% 15.7% 15.3%
Combined 0.8% 5.7% 8.4% 8.0%
Vent. Tach.
Training 0.3% 1.4% 41.8% 41.2%
Testing 0% 1.7% 55.2% 54.2%
Combined 0.2% 1.6% 48.0% 47.2%
All
Training 0.2% 26.0% 40.3% 29.9%
Testing 0.9% 22.1% 51.1% 39.8%
Combined 0.5% 24.0% 45.1% 34.3%
4.3 Performance of PPG-Based False Alarm Sup-
pression
The performance of the PPG-based false alarm suppression algorithm applied to the
training and test sets is summarized in Table 4.3. Across the entire collection of
alarms in both the training and test sets, the PPG-based algorithm suppressed 0.5%
of the true alarms and 24.0% of the false alarms.
On the training set, the algorithm suppressed true alarms only in the ventricular
tachycardia category. However in the test set, the PPG-based system suppressed true
alarms in two categories: asystole and extreme tachycarida. The highest rate of true
alarm suppression was in the asystole category. In the test set, 9.5% (or 2 of 21) of true
asystole alarms were suppressed. In both of these cases the PPG waveform exhibited
mid-frequency wave-like artifacts (most likely due to motion) but were not marked as
poor signal quality using the pSQI algorithm and asystole thresholds chosen in Table
68
4.2. The aP P G algorithm falsely detected pulses in these segments. This behavior
may be addressed by raising the noise floor in the aP P G algorithm.
The false alarm suppression performance of the PPG-based system varies widely
from one alarm category to another. Across both training and test sets, the algorithm
suppressed 68.2% of false asystole alarms, but only 1.7% of false ventricular tachy-
cardia alarms. The algorithm also suppressed only 1.6% of false extreme tachycardia
alarms in the test set. The low false alarm suppression rate in this category is often
due to poor signal qualtiy or heart rate estimates which exceeded the threshold of the
false alarm suppression algorithm. Raising the ventricular tachycardia rate threshold,
RV T , to 100 bpm raised the combined false alarm suppression rate to 5.7% but also
increased the true alarm suppression rate to 3.1%.
4.4 Limitations and Possible Improvements
The method of splitting our data into training and test sets was designed to keep the
number of patients equal for each alarm type. The poorer performance on the test set
indicates an asymmetry in the quality of signals and number of alarms investigated
in the training and test sets, with some circumstances arising in the test set which
did not appear in the training data. While the relative frequency of alarm types in
this study were similar in the training and test sets (see Tables 2.2 through 2.4), the
total alarm counts were unequal, illustrating that patient records do not contribute
equally to each alarm category and greatly in the number of alarms triggered. If
similar studies are to be performed in the future, we recommend that five fold cross-
validation is used to compensate for this imbalance.
Using the methods of PPG signal quality assessment and pulse onset detection
described in Chapters 2 and 3 of this thesis, the false ECG alarm suppression frame-
work using the PPG waveform does not perform as well as a similar framework which
utilizes the ABP waveform. The algorithm suppressed fewer false alarms than the
ABP-based system of Aboukhalil et al in all alarm categories and only had a large
impact for asystole alarms. Moreover, non-zero true alarm suppression rates are
69
probably clincally unacceptable for the asystole category. Fewer than 30% of false
alarms were suppressed in the case of extreme bradycardia, though no true alarms
were suppressed. The extreme tachycardia and ventricular tachycardia categories had
false alarm suppression rates as low as 5.7% and 1.6%, respectively, making the algo-
rithm of marginal use for these alarm types. Therefore it is hard to recommend the
approach described here for any alarm cateogories other than asystole and extreme
bradycardia. Even in the case of asystole, parameters may need further refinement
to reduce the true alarm suppression rate from 4% down to almost 0%.
The correlation between noise in the ECG channels and noise in the PPG waveform
has not been investigated. Artifacts in both channels may be due to the same source,
such as patient movement. This is a fundamental limitation to the PPG-based false
alarm suppression approach, especially because the PPG signal is highly sensitive to
movement. Correlations between ECG noise and PPG noise, as well as corellations
between detected ECG and PPG beats should be examined.
The false alarm suppression algorithm could not be evaluated in the case of ven-
tricular fibrillation because there were no instances of true ventricular fibrillation
alarms in the training or test data, and the ventricular fibrillation/tachycardia and
ventricular tachycardia alarms were combined into one category.
The components of the PPG-based false alarm suppression framework could be
improved and tuned in several ways. The windowing parameters and heart rate
thresholds used by the ABP-based false alarm suppression system of Aboukhalil et
al. [1] should be varied to find the optimal settings for this data set. In particu-
lar, the heart rate thresholds for tachycardia and ventricular tachycardia should be
investigated.
Extreme bradycardia and extreme tachycardia alarms are issued with the heart
rate threshold used by the monitor. The false alarm suppression rates for extreme
bradycardia and extreme tachycardia could be improved by using this heart rate
threshold as an input parameter to the aP P G algorithm. This would adjust the
refractory period of the pulse onset detector, and may improve heart rate esitmation
in these two alarm categories. However the method assumes the alarm is true, and
70
may not work if the heart rate greatly differs from the threshold on the monitor.
Like the ABP-based algorithm, the PPG-based false alarm suppression framework
has particular difficulty with examining periods of unsustained ventricular tachycar-
dia. This arrhythmia occurs if there is a ventricular rhythm (that is, one consisting
of beats which originate in the ventricles rather than from the sinoatrial node) with
a rate over 100 beats per minute, and is diagnosed using the electrocardiogram wave-
form. At low heart rates, the ABP and PPG waveforms cannot be used to infer these
details of the morphology of the ECG QRS complexes, and can only show the presence
of a beat. If the rate of the tachycardia is low enough (for example, 100 130 bpm),
the morphologies of the ABP or PPG pulses corresponding to these ventricular beats
appear to be normal. If in the future a similar logical structure is to be used for false
ventricular tachycardia alarm suppression, we recommend that the RV T heart rate
threshold be adjusted to increase the number of false alarms suppressed. However,
this may result in a larger number of true alarm suppressions.
71
THIS PAGE INTENTIONALLY LEFT BLANK
72
Chapter 5
Conclusions
5.1 Summary
5.1.1 Contributions
In this thesis we examined an algorithm which analyzes the PPG waveform for infor-
mation to use in the suppression of false critical ECG arrhythmia alarms issued by
ICU bedside monitors. To meet this goal, a signal quality metric, pSQI, and a pulse
onset detection algorithm, aP P G, were created to evaluate the fidelity of the PPG
signal and to assist in heart rate estimation.
A database of 4,326 annotated critical life-threatening ECG alarms was created to
tune the pSQI algorithm thresholds and evaluate the PPG-based false alarm suppres-
sion system. For benchmarking the performance of this new false alarm suppression
framework, the ABP-based algorithm of Aboukhalil et al. [1] was corrected to check
for low-volume pulses in the presence of extreme tachycardia, ventricular tachycardia,
or ventricular fibrillation alarms.
5.1.2 Evaluation and Limitations
Annotations of the alarms were made by individuals who were not involved in the cre-
ation of the ECG arrhythmia detection algorithms used in the ICU bedside monitors.
ECG alarms were therefore annotated using clinical criteria, and our alarm definitions
73
may not be completely consistent with the logic used by the monitoring algorithms
in some cases. The annotations for the “ventricular fibrillation/tachycardia” alarm
category in particular may not be consistent. The monitor algorithm fired an alarm
upon the detection of rapid ventricular tachycardia. The clinical annotator lableled
rapid ventricular tachycardia as “ventricular tachycardia” and the ventricular fibril-
lation annotation was reserved for asynchronous ventricular electrical activity. The
database did not show any true ventricular fibrillation events.
This annotation methodology affects the performance of the PPG signal quality
algorithm, pSQI . We quantified the spectral characteristics of the PPG waveform
using Hjorth parameters. PPG waveform segments corresponding to true alarms
in each category exhibit different heart rates and morphologies, and therefore have
different spectral characteristics. The Hjorth parameter thresholds for determining
the quality of the PPG signal are therefore set depending on the alarm type. The
Hjorth parameter thresholds chosen for ventricular tachycardia alarms in this study
will not provide the best assessment of signal quality under the condition of true
ventricular fibrillation.
The aP P G algorithm is limited in its ability to detect low-volume pulses. Recall
that the algorithm has an adaptive pulse confirmation threshold, which updates to
Tc% of the recent pulse amplitude when a pulse is detected, and decreases at a rate of
dc% per sample point (or Fs·dc% per second, where Fsis the sampling frequency) when
a pulse is not detected within 4 refractory periods. To avoid false pulse detections
due to low-amplitude artifacts, a noise floor is set at half of the smallest expected
true pulse amplitude. If the threshold is decreased below the noise floor, it is reset
to a value based on the recent amplitude of the PPG waveform. The rate at which
the pulse confirmation threshold adapts (dc) after a pulse has not been detected for
some time, and the value relative to the recent waveform amplitude (Tc) to which
that threshold adapts when a pulse is detected, have not been rigorously evaluated
during different arrhythmias. A trade-off in the noise floor setting exists when we
wish to detect pulses at different heart rates. In particular, if we wish to detect
low-amplitude pulses in the PPG waveform which appear during periods of extreme
74
tachycardia or ventricular tachycardia, the noise floor should be set at a low value to
increase sensitivity of the pulse detection algorithm. However, in the case of asystole
alarms we wish to decrease the sensitivity of the algorithm to artifactual oscillations
in the waveform to prevent false pulse detections.
The PPG-based false alarm suppression system does not perform as well as the
ABP-based suppression system with similar logic when both algorithms were applied
to the same set of annotated ECG alarms. The PPG-based algorithm performed
best in the asystole category, suppressing 68.2% of false asystole alarms and 4.0% of
true asystole alarms over both the training and test sets. False asystole alarms, false
extreme tachycardia alarms, and false ventricular tachycardia alarms comprise the
bulk of the false alarms in our data set. However, non-zero true alarm suppression
exists for the asystole, extreme tachycardia, and ventricular tachycardia categories, as
well as low false alarm suppression rates in the extreme tachycardia and ventricular
tachycardia categories. This indicates that if the asystole true alarm suppression
rate can be reduced 4.0% to 0% this algorithm will be of use in false asystole alarm
suppression.
5.2 Future Work
The application of the aP P G and pSQI algorithms in the false alarm suppression
framework highlighted several improvements which could be made to the pulse onset
and artifact detectors.
5.2.1 pSQI Improvement
The artifact detector uses a short window length in which to assess the quality of the
PPG waveform. In this study we have used 2 s non-overlapping windows in order
to make the artifact detector sensitive to the sudden onset of a disruptive artifact.
Overlapping the windows may decrease the sensitivity of the artifact detector to
these sudden artifacts. The effect of the choice of window size on the accuracy of the
Hjorth parameter estimates and the ability to detect artifacts should be evaluated.
75
The generation of artificial PPG artifacts, in a manner similar to Li et al. [16] may
be useful in conducting a controlled study of PPG artifact detection.
In this study, the Hjorth parameters were estimated from 30 s of PPG data pre-
ceding the onset of each ECG arrhythmia alarm. When this method for training the
Hjorth parameter thresholds was used, the false alarm suppression algorithm often
failed where the ECG exhibited an intermittent, non-sustained arrhythmia. For a
more accurate description of the range of Hjorth parameter values under each alarm
condition, the Hjorth parameters should be calculated from data where the arrhyth-
mia is present for the entire duration of the segment, such as 10 seconds surrounding
the alarm.
The distributions of H1and H2parameters illustrated in 2-4 and 2-5 are assumed
to take most of their “mass” from clean PPG waveforms. However, this assumption
was never tested. The alarm annotations were based on the morphology of the ECG
waveform. In future work, PPG segments with Hjorth parameters which lie within the
interquartile range should be examined to ensure that the PPG waveform morphology
is characteristic of the issued alarm.
Several settings of Hjorth parameter thresholds were investigated for each alarm
type to determine the proper signal quality thresholds for each arrhythmia. Fig-
ures 4-2 through 4-5 illustrate the effect of the Hjorth parameter threshold settings
on the performance of the PPG-based false alarm suppression algorithm. In some
alarm categories, the set of thresholds which yielded maximum false alarm suppres-
sion while minimizing true alarm suppression is very small. The range of the η1uand
η2thresholds tested should be extended until the false alarm suppression rate levels
off, especially in the extreme bradycardia alarm category.
As an alternative to a frequency-based signal quality metric, the use of dynamic
time warping of a matched filter or adaptive template could be evaluated, using cross-
correlation to quantify the extent to which each detected pulse deviates from normal
pulse morphology. Rapid non-physiological oscillations in oxygen saturation could
also serve as an estimate of signal quality in the PPG waveform; if oscillations in the
corresponding oxygen saturation estimates occur which are too rapid, the underlying
76
PPG waveform is likely to have been corrupted by artifact. However, the oxygen
saturation time series available to us may be too heavily filtered for this purpose.
Analysis using the raw waveform data from the pulse oximeter should be conducted
if possible, rather than using the signal which has been post-processed by the bedside
monitor or pulse oximeter module.
5.2.2 aP P G Improvement
A rigorous evaluation of the parameters of the pulse onset detector should be con-
ducted. For this evaluation, a database of annotated PPG pulse onsets should be
created, rather than using beat onsets from other waveforms, as was conducted in
this study.
Various settings of the threshold adaptivity rate, dc, and the threshold saturation
rate, Tc, should be performed on clean PPG waveforms exhibiting a range of heart
rates. A study of the sensitivity of the pulse onset detector to the noise floor setting
should be conducted.
Within the application of the aP P G algorithm to false ECG alarm suppression,
we should consider using the heart rate or alarm type reported by the monitor as an
input parameter to the aP P G algorithm. The refractory period in the aP P G algo-
rithm can be made longer or shorter based on the monitor’s heart rate threshold for
extreme bradycardia and extreme tachycardia alarms, which may improve pulse onset
detection performance. It may be the case that if the pulse threshold adapts more
quickly and the noise floor is lowered under tachycardia conditions, low-amplitude
pulses may be detectable. Similarly, if the noise floor is raised under asystole condi-
tions, false PPG pulse detections may prevent a true alarm from being suppressed.
However, this method risks decreasing the performance of the aP P G algorithm if the
alarm is false and the heart rate estimates from the ECG monitor and PPG waveform
are very different.
Future uses of the aP P G algorithm include a statistical analysis of the PPG pulse
amplitude and pulse duration, pulse transit time, and comparisons to the morphology
of the ECG and ABP activity immediately before any particular PPG pulse. That
77
is, if all are abnormal in a consistent manner, this may indicate the existence of an
arrhythmia.
5.2.3 False Alarm Suppression Improvement
The logic used to suppress false asystole and ventricular fibrillation alarms should be
improved. In both of these cases, the PPG waveform should not exhibit pulses, since
the heart is not pumping. Therefore, the same logic could be employed to suppress
false alarms in these categories.
The power spectrum of a non-pulsatile waveform does not exhibit any domi-
nant frequencies, and is instead characterized by the frequency components of noise.
Thresholding the power spectrum of PPG waveform segments which accompany true
asystole alarms does not differentiate between noise and a non-pulsatile PPG sig-
nal. The alarms should instead be suppressed if a pulsatile waveform is detected.
When assessing signal quality, rather than setting wide thresholds on the Hjorth pa-
rameters to check for a non-pulsatile waveform, the power spectrum should have a
dominant frequency estimate (H1) which lies between the Hjorth parameter ranges of
true bradycardia to true tachycardia. If the power spectrum exhibits characteristics
of a pulsatile signal, and pulses are detected in the PPG waveform, then the asystole
or ventricular tachycardia alarm should be suppressed.
False alarm suppression could be improved by combining the PPG-based frame-
work with the blood pressure framework where both signals are available. The combi-
nation of these two systems could be as simple as taking the output of the ABP-based
false alarm suppression system where it is available. Alternatively, the outputs could
be combined depending on the alarm type. For example, if either the ABP or PPG
waveforms contain detectable pulses in the presence of an asystole alarm, the alarm
may be marked as false.
The PPG signal quality framework could also be integrated into the robust es-
timation framework described by Li et al. and Nemati et al. [15, 16, 21]. This
framework uses signal quality metrics and Kalman flter-based innovation metrics to
quantify the relationships between simultaneously recorded ECG, ABP, respiration,
78
and PPG waveforms for continuous, robust heart rate, blood pressure, and respiration
estimation. Li et al. [16] have shown that fusing heart rate estimates from ECG and
ABP can improve the false alarm suppression rates for bradycardia and tachycardia.
It would be a simple extension to add the heart rate derrived from the PPG waveform
into this framework.
5.2.4 Other applications
A combination of PPG signal quality analysis and pulse onset detection could aid in
a study on detection of dampening in the ABP waveform. These algorithms could
also be applied to a study of the detection of vasoconstriction, by examining the PPG
pulse amplitude and the pulse transit time between each ABP pulse onset detection
and the corresponding PPG pulse onset detection.
5.3 Extensibility
The false alarm suppression algorithm presented in this thesis applies as an extension
of multi-state system tracking methods being researched at the Charles Stark Draper
Laboratory. As part of a space technology program, Draper has been working on the
Multi-State Excursion Assessment (MSEA) algorithm, a generic system monitoring
and state enunciation algorithm. Originally developed for monitoring the “health” of
a vehicle trajectory, the MSEA algorithm monitors multiple vehicle parameters and
states, such as position, altitude, and velocity to determine if the landing objective is
attainable or unattainable. Nominal values and tolerable error margins are provided
for each monitored state and regularly updated. Note that the MSEA algorithm does
not identify corrections to undesired states.
The MSEA algorithm could also be applied to physiologic monitoring. In this
setting, states might include fiducial markers in physiologic waveforms (such as QRS
complexes or pulse onsets) and derived parameters may include heart rate or blood
oxygen saturation estimates. Such an envelope system for physiologic monitoring
may behave similarly to monitoring technology found in implantable medical devices
79
or bedside monitors found in hospital ICUs.
The MSEA algorithm relies on the assumption that state information received
from sensor inputs is valid and not corrupted. In the presence of corrupt inputs, the
MSEA algorithm might return a false or invalid assessment of whether the landing
objective is attainable. The MSEA algorithm could therefore benefit from an enve-
lope system which assesses the validity of the received state estimates and determines
if the MSEA algorithm output is true. The physiologic monitoring MSEA algorithm
extension may behave similarly to the PPG signal quality assessment components of
the alarm suppression algorithm presented in this thesis. The envelope system for
trajectory assessment might instead perform this verification using mutual informa-
tion between related states, for example to estimate velocity from previously recorded
positions.
80
Appendix A
False ECG Alarm Suppression
Using the ABP Waveform
Clifford et al. and Aboukhalil et al. have created a logical framework to suppress false
heart-rate related alarms issued by ECG bedside monitors in ICU settings [3, 1]. The
logic and reported performance of this false alarm suppression framework is briefly
reviewed in this appendix. For benchmarking of the PPG false alarm suppression
framework described in Chapter 4, the ABP framework was also applied to the data
set described in Sections 2.3.1 and 2.3.2. The performance on this new data set are
also included in this appendix.
A.1 Original Algorithm Architecture
At the onset of each critical ECG arrhythmia alarm, a 17-second ABP waveform
segment is extracted from 13 seconds prior the alarm to 4 seconds after. Alarms
where the blood pressure waveform was not available for this 17-second window were
excluded from the study. Next, the wABP algorithm [32] is applied to determine
the onset time of each pulse in the ABP waveform, and the signal quality index of
each detected beat is calculated using the negated output of the jSQI algorithm [26].
At the end of this processing step, each detected beat has been marked as either
good (high signal quality) or abnormal (low signal quality). If fewer than 80% of the
81
detected ABP beats are marked as high signal quality, the ABP signal quality for
the 17 s segment is judged to be poor and the ECG alarm is accepted as true. If the
signal quality for the segment is high, the logic employed to assess the validity of the
alarm depends on the alarm type. Figure A-1 illustrates the algorithm architecture
where the ABP signal quality is sufficiently high. The heart rate parameters found
to be optimal by Aboukhalil et al. are summarized in Table A.3.
Figure A-1: False ECG Alarm Suppression Using the ABP Waveform. If at least 80%
of the beats detected in a 17-second window are of high signal quality, logic is used to
determine if the alarm should be accepted or suppressed. “IPI” stands for inter-pulse
interval. “SAI” stands for signal abnormality index. Figure adapted from [1].
A.1.1 Asystole Processing
From the detected beat onsets, the largest beat-to-beat interval (for the case where the
asystole resolves itself within the window) and the interval between the last detected
pulse and the end of the window (for the case where asystole lasted beyond the analysis
period) are calculated. If either of these intervals exceeds TA= 3 s, the absence of
beats is noted and the asystole alarm is accepted; otherwise it is suppressed.
82
A.1.2 Extreme Bradycardia Processing
Bradycardia alarms in the ECG waveforms are marked relative to a heart rate thresh-
old set on the monitor. The NB= 3 longest pulse-to-pulse intervals are calculated.
If the mean heart rate calculated from these beats is within EB= 7 beats per minute
of the monitor’s heart rate, then the bradycardia alarm is accepted; otherwise it is
suppressed.
A.1.3 Extreme Tachycardia Processing
Tachycardia alarms in the ECG waveforms are marked relative to a heart rate thresh-
old set on the monitor. The NT= 1 shortest pulse-to-pulse interval is used to calculate
the heart rate. If the heart rate estimate is less than ET= 20 beats per minute over
the threshold set on the monitor, and there are fewer than MT= 5 abnormal beats
lasting less than TT= 4 s (indicating no period of abnormal beat morphology), then
the tachycardia alarm is suppressed.
A.1.4 Ventricular Tachycardia Processing
A ventricular tachycardia alarm is accepted if the total duration of abnormal detected
beats is more than TV T = 2 s (indicating abnormal beat morphology) and the heart
rate calculated from the NV T = 1 shortest pulse-to-pulse interval exceeds RV T = 80
beats per minute. Otherwise, the alarm is suppressed.
A.1.5 Ventricular Fibrillation Processing
Ventricular fibrillation alarms are suppressed if the total duration of abnormal de-
tected beats is less than TV F = 2.5 s (indicating no period of abnormal beat morphol-
ogy) and the heart rate calculated from the NV F = 1 shortest pulse-to-pulse interval
is less than RV F = 150 beats per minute. Otherwise, the alarm is accepted.
83
Table A.1: Performance of ABP-based false alarm suppression algorithm reported by
Aboukhalil et al. The training set contained 267 alarms, and the test set contained
180 alarms. “FA” signifies false alarm. “TA” signifies true alarm. “FA Rate Before”
and “FA Rate After” refer to the false alarm rate before and after the suppression
algorithm was used, respectively. “-” indicates a value which was not reported.
Alarm Type Data Set TA FA FA Rate FA Rate
Suppression Suppression Before After
Asystole
Training 0% 92.5% - -
Testing 0% 95.0% - -
Combined 0% 93.5% 90.7% 5.5%
Ext. Brad.
Training 0% 79.7% - -
Testing 0% 83.6% - -
Combined 0% 81.0% 29.3% 5.5%
Ext. Tach.
Training 0% 59.4% - -
Testing 0% 70.1% - -
Combined 0% 63.7% 23.1% 8.4%
Vent. Tach.
Training 14.5% 28.3% - -
Testing 4.0% 38.7% - -
Combined 9.4% 33.0% 46.6% 30.8%
Vent. Fib.
Training 0% 57.7% - -
Testing 0% 58.9% - -
Combined 0% 58.2% 79.6% 33.1%
All
Training 4.7% 57.0% - -
Testing 1.4% 63.2% - -
Combined 2.4% 59.7% 42.7% 17.2%
A.1.6 Performance on Unseen Data
The performance of the false ECG alarm suppression system described above, as
reported by Aboukhalil et al., is summarized in Table A.1. This ABP-based system
was also applied to the data described in Sections 2.3.1 and 2.3.2. The performance
of the algorithm on the new MIMIC II data is summarized in Table A.2.
The ABP-based algorithm performed similarly on new data from the MIMIC II
database as reported in the study of Aboukhalil et al.. The false alarm rate after ap-
plying the original ABP framework was higher on our data set for asystole alarms, as
fewer false alarms were suppressed in this category than reported in the Aboukhalil’s
study. The false alarm suppression rate was also lower for extreme tachycardia and
ventricular tachycardia alarms in our data set, and true alarm suppression rates were
84
Table A.2: Performance of ABP-based false alarm suppression algorithm on new
MIMIC II data (described in Sections 2.3.1 and 2.3.2). Note that ventricular fibrilla-
tion/tachycardia alarms are combined with and annotated as ventricular tachycardia
alarms. “FA” signifies false alarm. “TA” signifies true alarm. “FA Rate Before”
and “FA Rate After” refer to the false alarm rate before and after the suppression
algorithm was used, respectively.
Alarm Type Data Set TA FA FA Rate FA Rate
Suppression Suppression Before After
Asystole
Training 0% 70.9% 91.0% 26.5%
Testing 0% 53.3% 90.0% 42.0%
Combined 0% 62.7% 90.5% 33.8%
Ext. Brad.
Training 0% 90.6% 35.0% 3.3%
Testing 0% 50.0% 16.1% 8.0%
Combined 0% 81.7% 27.8% 5.1%
Ext. Tach.
Training 1.0% 54.6% 2.6% 1.2%
Testing 0% 15.0% 12.9% 11.0%
Combined 0.7% 29.0% 5.3% 3.8%
Vent. Tach.
Training 5.0% 26.7% 40.4% 29.6%
Testing 10.8% 35.8% 54.9% 35.2%
Combined 8.0% 31.9% 47.6% 32.4%
All
Training 3.2% 48.1% 36.9% 19.2%
Testing 6.6% 40.2% 52.0% 31.1%
Combined 4.5% 44.0% 43.5% 24.4%
85
non-zero for both extreme tachycardia and ventricular tachycardia (which was not
observed in Aboukhalil’s study).
A.1.7 Limitations
The ABP-based false alarm suppression system described by Aboukhalil et al. [1] had
the poorest false alarm suppression rate (31.9%) on ventricular tachycardia alarms.
Alarms in this category were also the only true alarms suppressed in the original
study. The authors attribute the true ventricular tachycardia alarm suppressions
to the observation that at low heart rates during ventricular tachycardia, the ABP
morphology looks quite normal. Since the algorithm accepts ventricular tachycardia
alarms if the calculated heart rate is over RV T = 80 beats per minute and the total
duration of detected abnormal pulses is more than TV T = 2 s (indicating abnormal
beat morphology), true ECG ventricular tachycardia alarms are suppressed when the
corresponding ABP beats have normal morphology.
On our data set, the ABP-based algorithm also suppressed true extreme tachycar-
dia alarms. In the ABP segments surrounding the suppressed true alarms, the pulses
exhibiting rapid instantaneous heart rate have low pulse pressure and are not de-
tected by wABP . The heart rate estimated from the detected pulses is therefore less
than RT= 80 beats per minute. The last detected pulse preceding the low-amplitude
section is typically marked as having a long duration, but is the only abnormal beat
(therefore the number of abnormal beats is less than MT= 5). The duration of this
one abnormal beat is less than TT= 4 s. Because all three of these conditions are met
from a non-sustained period of extreme tachycardia, the true tachycardia alarms are
suppressed. This could be avoided either by introducing a threshold on the duration
of the longest inter-pulse interval, or by reducing the number of abnormal beats or
total duration of abnormal beats to account for unsustained episodes of tachycardia.
86
A.2 Modifications Made for Benchmarking
We used the performance of this ABP-based algorithm as a benchmark for the perfor-
mance of the PPG-based false alarm suppression algorithm developed in this thesis.
In order to reduce true alarm suppression, a modification has been made to the
ABP-based algorithm. For each the three alarms related to fast heart rates (extreme
tachycardia and ventricular tachycardia, and ventricular fibrillation), a new condi-
tion has been introduced regarding the longest inter-pulse interval. If the longest
inter-pulse interval exceeds TN oP ulse = 1.8 s then the waveform indicates a tempo-
rary loss of pulse due to low stroke volume, resulting in some number of missed beat
detections. The resulting algorithm architecture is as follows. Extreme tachycardia
alarms are suppressed if the longest inter-pulse interval is less than TN oP ulse, the heart
rate calculated from the shortest inter-pulse interval is less than ET= 20 beats per
minute over the monitor’s declared heart rate threshold, and fewer than MT= 5
abnormal beats lasting fewer than TT= 4 s are detected. Ventricular tachycardia
alarms are accepted if either the longest inter-pulse interval is greater than TN oP ulse
or both the total duration of abnormal detected beats is more than TV T = 2 s and
the heart rate calculated from the shortest pulse-to-pulse interval exceeds RV T = 80
beats per minute. Ventricular fibrillation alarms are suppressed if the longest inter-
pulse interval is less than TN oP ulse , the total duration of abnormal detected beats is
less than TV F = 2.5 s, and the mean heart rate calculated from the NV F = 1 shortest
pulse-to-pulse interval is less than RV F = 150 beats per minute. The windowing and
thresholding parameters are summarized in Table A.3.
A.2.1 Performance on Unseen Data
This modified system was applied to the data described in Sections 2.3.1 and 2.3.2.
The performance of this modified ABP-based algorithm is summarized in Table A.4.
Modifying the ABP-based alarm suppression framework to detect loss in pulse
volume decreased the true alarm suppression rate for extreme tachycardia alarms in
the training set from 1.0% to 0.2% but increased the true alarm suppression rate in
87
Table A.3: Windowing and thresholding parameters in merged ABP-based false alarm
suppression algorithm. “PPI” signifies pulse-to-pulse interval. “N/A” signifies that
the parameter is not applicable.
Parameter Max. PPI HR error Intervals Duration Abnormal Max. HR
length (s) margin for HR of bad beats (bpm)
Alarm Type (bpm) calc. beats (s) allowed
Asystole TA= 3 N/A N/A N/A N/A N/A
Extreme N/A EB= 7 NB= 3 N/A N/A N/A
Bradycardia
Extreme TNoP ulse = 1.8ET= 20 NT= 1 TT= 4 MT= 5 N/A
Tachycardia
Ventricular TNoP ulse = 1.8 N/A NV T = 1 TV T = 2 N/A RV T = 80
Tachycardia
Ventricular TNoP ulse = 1.8 N/A NV F = 1 TV F = 2.5 N/A RV F = 150
Fibrillation
Table A.4: Performance of modified ABP-based false alarm suppression algorithm
on new MIMIC II data (described in Sections 2.3.1 and 2.3.2). Note that ventric-
ular fibrillation/tachycardia alarms are combined with and annotated as ventricular
tachycardia alarms. “FA” signifies false alarm. “TA” signifies true alarm. “FA Rate
Before” and “FA Rate After” refer to the false alarm rate before and after the modified
suppression algorithm was used, respectively.
Alarm Type Data Set TA FA FA Rate FA Rate
Suppression Suppression Before After
Asystole
Training 0% 70.9% 91.0% 26.5%
Testing 0% 53.3% 90.0% 42.0%
Combined 0% 62.7% 90.6% 33.8%
Ext. Brad.
Training 0% 90.6% 35.0% 3.3%
Testing 0% 50.0% 16.1% 8.0%
Combined 0% 81.7% 27.8% 5.1%
Ext. Tach.
Training 0.2% 54.6% 2.6% 1.2%
Testing 0% 15.0% 12.9% 11.0%
Combined 0.2% 29.0% 5.3% 3.8%
Vent. Tach.
Training 5.8% 25.2% 40.4% 30.2%
Testing 10.4% 35.8% 54.9% 35.2%
Combined 7.8% 31.3% 47.6% 32.7%
All
Training 2.9% 47.3% 36.9% 19.5%
Testing 6.3% 40.2% 52.0% 31.1%
Combined 4.2% 43.6% 43.5% 24.6%
88
the ventricular tachycardia category from 5.0% to 5.8%. The false alarm suppression
rate for ventricular tachycardia alarms in the training set decreased slightly from
26.7% to 25.2%.
89
THIS PAGE INTENTIONALLY LEFT BLANK
90
Bibliography
[1] A Aboukhalil, L Nielsen, M Saeed, R G Mark, and G D Clifford. Reducing false
alarm rates for critical arrhythmias using the arterial blood pressure waveform.
Journal of Biomedical Informatics, 41(3):442–451, 2008.
[2] M Chambrin. Review: Alarms in the intensive care unit: how can the number
of false alarms be reduced? Critical Care, 5(4):184–188, May 2001.
[3] G D Clifford, A Aboukalil, J X Sun, W Zong, B A Janz, G B Moody, and R G
Mark. Using the blood pressure waveform to reduce critical false ECG alarms.
In Computers in Cardiology, volume 33, pages 829–832, 2006.
[4] F M Coetzee and Z Elghazzawi. Noise-resistant pulse oximetry using a synthetic
reference signal. IEEE Transactions on Biomedical Engineering, 47(8):1018–
1026, August 2000.
[5] M Drinnan, J Allen, and A Murray. Relation between heart rate and pulse transit
time during paced respiration. Physiological Measurement, 22(3):425–432, 2001.
[6] R Flewelling. Noninvasive optical monitoring. In JD Bronzino, editor, The
Biomedical Engineering Handbook, pages 1346–1352. CRC Press, 1995.
[7] P T Gibbs, L B Wood, and H H Assada. Active Motion Artifact Cancellation for
Wearable Health Monitoring Sensors using MEMS Accelerometers. SPIE Smart
Structures and Materials, 5765:811–819, May 2005.
[8] E Gil, V Monasterio, P Laguna, and J M Vergara. Pulse photopletismography
amplitude decrease detector for sleep apnea evaluation in children. In 27th An-
nual International Conference of the IEEE Engineering in Medicine and Biology
Society, volume 3, pages 2743–2746, 2005.
[9] E Gil, J M Vergara, and P Laguna. Detection of decreases in the amplitude
fluctuation of pulse photoplethysmography signal as indication of obstructive
sleep apnea syndrome in children. Biomedical Signal Processing and Control,
3(3):267 – 277, February 2008.
[10] A L Goldberger, L A N Amaral, L Glass, J M Hausdorff, P Ch Ivanov, R G Mark,
J E Mietus, G B Moody, C-K Peng, and H E Stanley. PhysioBank, PhysioToolkit,
and PhysioNet: Components of a new research resource for complex physiologic
91
signals. Circulation, 101(23):e215–e220, 2000. Circulation Electronic Pages:
http://circ.ahajournals.org/cgi/content/full/101/23/e215.
[11] J Greenberg, J Fisher, W Wells, and G D Clifford. HST.582J / 6.555J / 16.456J
Biomedical Signal and Image Processing. (Massachusetts Institute of Technol-
ogy: MIT OpenCouseWare), http://ocw.mit.edu (Accessed July 2009). License:
Creative Commons BY-NC-SA, Spring 2007.
[12] M J Hayes and P R Smith. Quantitative evaluation of photoplethysmographic
artefact reduction for pulse oximetry. In EUROPTO Conference on Medical
Sensors and Fiber Optic Sensors IV, volume 3570, pages 138–147, September
1998.
[13] B Hjorth. EEG analysis based on time domain properties. Electroencephalography
and Clinical Neurophysiology, 29:306–310, 1970.
[14] B Hjorth. The physical significance of time domain descriptors in EEG analysis.
Electroencephalography and Clinical Neurophysiology, 34:321–325, 1970.
[15] Q Li, R G Mark, and G D Clifford. Robust heart rate estimation from multi-
ple asynchronous noisy sources using signal quality indices and a kalman filter.
Institute of Physics Physiological Measuremet, 29:15–32, 2008.
[16] Q Li, R G Mark, and G D Clifford. Artificial arterial blood pressure artifact
models and an evaluation of a robust blood pressure and heart rate estimator.
BioMedical Engineering OnLine, 8(1):13, 2009.
[17] S P Linder, S M Wendelken, E Wei, and S P McGrath. Using the morphology
of photoplethysmogram peaks to detect changes in posture. Journal of Clinical
Monitoring and Computing, 20(3):151–158, June 2006.
[18] R G Mark. HST.542J / 2.792J / 20.371J / 6.022J Quantitative Physiology:
Organ Transport Systems. (Massachusetts Institute of Technology: MIT Open-
CouseWare), http://ocw.mit.edu (Accessed July 2009). License: Creative Com-
mons BY-NC-SA, Spring 2004.
[19] G B Moody. WAVE user’s guide. Available online at
http://www.physionet.org/physiotools/wug/ (Accessed June 2009).
[20] G B Moody and R G Mark. A database to support development and evaluation
of intelligent intensive care monitoring. Computers in Cardiology, 23:657–660,
1996.
[21] S. Nemati and Cifford G. D. Data fusion for improved respiration rate estimation.
Technical report, Massachusetts Institute of Technology, January 2009.
[22] C F Poets and V A Stebbens. Detection of movement artifact in recorded pulse
oximeter saturation. European Journal of Pediatrics, 156(10):808–811, 1997.
92
[23] M Saeed, C Lieu, G Raber, and R G Mark. MIMIC II: a massive temporal ICU
patient database to support research in intelligent patient monitoring. Computers
in Cardiology, 29:641–644, 2002.
[24] J W Severinghaus and J F Kelleher. Recent developments in pulse oximetry.
Anesthesiology, 76(6):1018–1038, 1992.
[25] R Sokwoo, B H Yang, and H H Asada. Artifact-resistant power-efficient design of
finger-ring plethysmographic sensors. IEEE Trans Biomed Eng, 48(7):795–805,
2001.
[26] J X Sun, A T Reisner, and R G Mark. A signal abnormality index for arterial
blood pressure waveforms. Computers in Cardiology, 33:13–16, 2006.
[27] K K Tremper and S J Barker. Pulse oximetry. Anesthesiology, 70(1):98–108,
1989.
[28] L B Wood and H H Asada. Active motion artifact reduction for wearable sensors
using Laguerre expansion and signal separation. In 27th Annual International
Conference of the IEEE Engineering in Medicine and Biology Society, pages
3571–3574, January 2005.
[29] L B Wood and H H Asada. Low variance adaptive filter for cancelling motion
artifact in wearable photoplethysmogram sensor signals. In 29th Annual Inter-
national Conference of the IEEE Engineering in Medicine and Biology Society,
pages 652–655, August 2007.
[30] M Yarita, N Kobayashi, S Takeda, and T Tamura. Compensation for two specific
types of artifact in pulse wave using a Kalman filter. Information Technology
Applications in Biomedicine, 2007. ITAB 2007. 6th International Special Topic
Conference on, pages 269–272, 8-11 Nov. 2007.
[31] C Yu, Z Liu, T McKenna, A T Reisner, and J Reifman. A method for automatic
identification of reliable heart rates calculated from ECG and PPG waveforms.
Journal of the American Medical Informatics Association, 13(3):309–320, Jun
2006.
[32] W Zong, T Heldt, G B Moody, and R G Mark. An open-source algorithm to
detect onset of arterial blood pressure pulses. Computers in Cardiology, pages
259–262, September 2003.
[33] W Zong, G B Moody, and R G Mark. Reduction of false arterial blood pres-
sure alarms using signal quality assessment and electrocardiogram-arterial blood
pressure relationships. Medical and Biological Engineering and Computing, 42,
2004.
93
... Otero et al. [19] proposed a multivariable fuzzy temporal profile model which described a set of monitoring criteria of temporal evolution of the patient's physiological variables of HR, oxygen saturation (SpO 2 ) and BP. Aboukhalil et al. [14] and Deshmane [20] used synchronous ABP and PPG signals to suppress false ECG alarms. Zong et al. [21] reduced false ABP alarms using the relationships between ECG and ABP. ...
... Since the VT alarm was considered the most difficult type of false alarm to suppress, with an associated low false alarm reduction rate and high true alarm suppression rate in literature [14,[20][21][22][23]28], we therefore focus on reducing this false alarm for the rest of the chapter. Interested readers are directed to Li and Clifford [23] for methods to reduce false alarms on the other types of alarms. ...
... In total 147 features and SQI metrics were extracted from ECG, ABP, PPG, and SpO 2 signals within the 10 s analysis window. These features were generally chosen based upon previous research by the authors and others [14,[20][21][22][23][24][28][29][30][31][32]. The typical features included HR (extracted from ECG, ABP, and PPG), blood pressure (systolic, diastolic, mean), oxygen saturation (SpO 2 ), and the amplitude of PPG. ...
Chapter
Full-text available
Modern patient monitoring systems in intensive care produce frequent false alarms that can lead to reduced standard of care. In this case study we demonstrate the development of an algorithm which uses a data fusion and machine learning approach to reduce the number of false alarms, while avoiding the suppression of a significant number of true alarms.
... The pulse sensor works in the style of a photoplethysmography (PPT) [8]. When pumping blood throughout the body, with each beat there is a pulse wave traveling through all the arteries to the ends of the hair tissue such as the fingertips and earlobes. ...
Conference Paper
Full-text available
In the work environment there are numerous jobs that involve the performance of tasks in cold environments, such as storage in cold rooms. Knowing that temperature is an aspect that alters the heart rate, working in these conditions can generate risk factors that favor the appearance of pathologies, disorders and aggravate chronic diseases, such as cardiac, respiratory and rheumatic. In this project, we carry out a series of analytical stages that evaluate the cardiac activity of workers exposed continuously to very low temperatures. Through this reasoning, we provide information involved in the accurate interpretation of the electrical activity of the heart to detect abnormalities and determine risk factors that increase the chances of suffering physical or mental ills. As a result, we present the data obtained from 10 workers in 10 weeks, when performing the analysis, compare the graphs and data exposed by each one and determine the presence of anomalous signals. All this for the correct management, knowledge and evaluation, in order to avoid that the cold is another work risk faced by many workers.
... SQI ECG In clinical ECG analysis, a great deal of emphasis has been placed on the definition of SQI metrics to reduce false alarm rates in adult intensive care units (ICUs). With this in mind, most definitions of ECG signal quality found in the literature have been developed for adult ICU patient data, particularly in the work of [40][41][42]. In our analysis, in order to assess the quality of ECG records, the SQI defined in [41] was applied as originally implemented by the authors with the exception of the following: (a) when assessing the quality of the single-lead ECG signal bSQI (Section 2.1.1, ...
Article
Full-text available
Knowledge of the pathological instabilities in the breathing pattern can provide valuable insights into the cardiorespiratory status of the critically-ill infant as well as their maturation level. This paper is concerned with the measurement of respiratory rate (RR) in premature infants. We compare the rates estimated from (a) the chest impedance pneumogram, (b) the ECG-derived respiratory rhythms, and (c) the PPG-derived respiratory rhythms against those measured in the reference standard of breath detection provided by attending clinical staff during 165 manual breath counts. We demonstrate that accurate RR estimates can be produced from all sources for RR in the 40 - 80 bpm (breaths per min) range. We also conclude that the use of indirect methods based on the ECG or the PPG poses a fundamental challenge in this population due to their poor behaviour at fast breathing rates (upwards of 80 bpm).
... Hjorth parameters can be used to evaluate the signal quality and then this algorithm was improved by Deshmane [2][3][4][5]. He put forward an adaptive artifact detection algorithm to suppress the false electrocardiogram (ECG) alarms. ...
... First of all, PPG extracted through smartphone, pulse oximeter or wearable sensor is often corrupted with ambient noise, motion artifacts and other noise sources. Even the warning systems in hospital ICU (Intensive Care Unit) are prone to transient noise [2], which leads to high number of false alarms [2][3]. In fact, high false negatives (higher undetected cardiac conditions) would be deadly to the patients [4]. ...
Chapter
In this paper, we present CardioFit, a completely noninvasive cardiac condition monitoring system that enhances the clinical utility of health care analytics like lowering false detection of cardiac arrhythmia condition, higher accuracy in heart rate variability (HRV) computation. It performs powerful local analysis to enable accurate as well as easy-to-use, round-the-clock in-house, remote or mobile cardiac health checking. Here, photoplethysmogram (PPG) is the sole physiological signal considered for cardiac health management. It is to be noted that PPG carries significant necessary features what is available from electrocardiogram (ECG) signal. Unlike ECG, extraction of PPG is noninvasive, easy and affordable using smartphone or other low cost sensors. However, PPG is frequently contaminated with various kinds of motion artifacts and noise. Our robust concoction of signal processing and machine learning techniques exhibit higher accuracy in the detection and removal of the corrupt PPG signal segments. The proposed mechanism substantially improves the detection capability of the cardiac condition. Efficacy of our scheme is depicted using publicly available MIT-Physionet database as well as through our own field-collected real-life PPG data.
... In recent years, several strategies have been developed to tackle this problem. For example, Aboukhalil et al. (Aboukhalil et al 2008) and Deshmane (Deshmane 2009) applied a multi-parameter analysis on ECG and pulsatile waveforms to improve algorithms for alarm generation. However, both methods met the problem that alarms of ventricular tachycardia had high true alarm (TA) suppression rates while low FA reduction rates. ...
Article
To lessen the rate of false critical arrhythmia alarms, we used robust heart rate estimation and cost-sensitive support vector machines. The MIMIC database and the PhysioNet/Computing in Cardiology Challenge 2015 public dataset constituted the training set, and the Challenge 2015 hidden dataset was the test set. Each record had an alarm labeled with asystole, extreme bradycardia, extreme tachycardia, ventricular tachycardia or ventricular flutter/fibrillation. Before alarm onsets, 300-second multimodal data was provided, including electrocardiogram, arterial blood pressure, and/or photoplethysmogram. The signal quality modified Kalman filter enforced robust heart rate estimation. Based on that, we extracted heart rate variability features and ECG derived features. Next, we applied a genetic algorithm (GA) to select the optimal feature combination. Finally, considering the high cost of classifying a true arrhythmia as false, we selected cost-sensitive support vector machines (CSSVM) to classify alarms. Evaluation on the test set showed the overall true positive rate was 95%, and the true negative rate was 85%.
Chapter
Real-time monitoring of life-threatening cardiovascular disease like arrhythmia using wearable sensors and Internet of things (IoT) devices paves ways to mobile health (m-health) systems. Smartphones with developed applications, wearable sensors and IoT devices are the major parts of the developed real-time arrhythmia monitoring system. In this work, an in-house round-the-clock cardiac monitoring is proposed with the use of machine learning techniques to predict the symptoms of arrhythmia by classifying the data obtained from UCI repository. The physiological signal electrocardiogram (ECG) is considered to characterize the anomalous behavior of the cardiac system. Our main novelty is to predict the symptoms of arrhythmia with the analysis and classification of data obtained from the patients using sensors or smartphones to the data classified at the repository. We establish the accuracy and efficiency of the proposed solution, by analyzing the large set of data with the field collected ECG signals.
Conference Paper
Atrial fibrillation (AFib) is diagnosed by analysis of the morphological and rhythmic properties of the electrocardiogram. It was recently shown that accurate detection of AFib is possible using beat-to-beat interval variations. This raises the question of whether AFib detection can be performed using a pulsatile waveform such as the Photoplethysmogram (PPG). The recent explosion in use of recreational and professional ambulatory wrist-based pulse monitoring devices means that an accurate pulse-based AFib screening algorithm would enable large scale screening for silent or undiagnosed AFib, a significant risk factor for multiple diseases. We propose a noise-resistant machine learning approach to detecting AFib from noisy ambulatory PPG recorded from the wrist using a modern research watch-based wearable device (the Samsung Simband). Ambulatory pulsatile and movement data were recorded from 46 subjects, 15 with AFib and 31 non symptomatic. Single channel electrocardiogram (ECG), multi-wavelength PPG and tri-axial accelerometry were recorded simultaneously at 128 Hz from the non-dominant wrist using the Simband. Recording lengths varied from 3.5 to 8.5 minutes. Pulse (beat) detection was performed on the PPG waveforms, and eleven features were extracted based on beat-to-beat variability and waveform signal quality. Using 10-fold cross validation, an accuracy of 95 % on out-of-sample data was achieved, with a sensitivity of 97%, specificity of 94%, and an area under the receiver operating curve (AUROC) of 0.99. The described approach provides a noise-resistant, accurate screening tool for AFib from PPG sensors located in an ambulatory wrist watch. To our knowledge this is the first study to demonstrate an algorithm with a high enough accuracy to be used in general population studies that does not require an ambulatory Holter electrocardiographic monitor.
Article
Full-text available
This paper presents an active noise cancellation technique for recovering wearable biosensor signals corrupted by bodily motion. A finger mounted photoplethysmograph (PPG) ring sensor with a collocated MEMS accelerometer is considered. The system by which finger acceleration disturbs PPG output is identified and a means of modeling this relationship is prescribed using either FIR or Laguerre models. This means of modeling motivates the use of a recursive least squares active noise cancellation technique using the MEMS accelerometer reading as an input for a FIR or Laguerre model. The model parameters are identified and tuned in real time to minimize the power of the recovered PPG signal. Experiments show that the active noise cancellation method can recover pulse information from PPG signals corrupted with up to 2G of acceleration with 85% improvement in mean squared error.
Conference Paper
Full-text available
Development and evaluation of intensive care unit (ICU) decision-support systems would be greatly facilitated by the availability of a large-scale ICU patient database. Following our previous efforts with the MIMIC (Multi-parameter Intelligent Monitoring for Intensive Care) Database, we have leveraged advances in networking and storage technologies to develop a far more massive temporal database, MIMIC II. MIMIC II is an ongoing effort: data is continuously and prospectively archived from all ICU patients in our hospital. MIMIC II now consists of over 800 ICU patient records including over 120 gigabytes of data and is growing. A customized archiving system was used to store continuously up to four waveforms and 30 different parameters from ICU patient monitors. An integrated user-friendly relational database was developed for browsing of patients' clinical information (lab results, fluid balance, medications, nurses' progress notes). Based upon its unprecedented size and scope, MIMIC II will prove to be an important resource for intelligent patient monitoring research, and will support efforts in medical data mining and knowledge-discovery.
Conference Paper
Full-text available
Intensive Care Unit (ICU) false alarm rates can be as high as 86%, leading to a desensitization of the clinical attending staff, slowing of response times and even ignoring true alarms. False alarms are commonly caused by single channel artifacts and could be avoided if information from other independent signals were fused to form a more robust hypothesis of the etiology of the alarm. We used a standard multi-parameter ICU database (PhysioNet's MIMIC DB) to investigate the frequency of false critical (or 'life- threatening') arrhythmia alarms produced by a commercial ICU monitoring system. Multiple expert reviews of the alarms were made using all the relevant files in the MIMIC DB (a total of 21 subjects and 800 hours of waveform data). We found that 25% of the 89 life-threatening alarms were considered false. We then implemented an algorithm to suppress false alarms, using information derived from the arterial blood pressure signal. This simple yet robust strategy was successful in suppressing all false alarms, without suppressing any true alarms.
Article
Full-text available
Within the intensive care unit (ICU), arterial blood pressure (ABP) is typically recorded at different (and sometimes uneven) sampling frequencies, and from different sensors, and is often corrupted by different artifacts and noise which are often non-Gaussian, nonlinear and nonstationary. Extracting robust parameters from such signals, and providing confidences in the estimates is therefore difficult and requires an adaptive filtering approach which accounts for artifact types. Using a large ICU database, and over 6000 hours of simultaneously acquired electrocardiogram (ECG) and ABP waveforms sampled at 125 Hz from a 437 patient subset, we documented six general types of ABP artifact. We describe a new ABP signal quality index (SQI), based upon the combination of two previously reported signal quality measures weighted together. One index measures morphological normality, and the other degradation due to noise. After extracting a 6084-hour subset of clean data using our SQI, we evaluated a new robust tracking algorithm for estimating blood pressure and heart rate (HR) based upon a Kalman Filter (KF) with an update sequence modified by the KF innovation sequence and the value of the SQI. In order to do this, we have created six novel models of different categories of artifacts that we have identified in our ABP waveform data. These artifact models were then injected into clean ABP waveforms in a controlled manner. Clinical blood pressure (systolic, mean and diastolic) estimates were then made from the ABP waveforms for both clean and corrupted data. The mean absolute error for systolic, mean and diastolic blood pressure was then calculated for different levels of artifact pollution to provide estimates of expected errors given a single value of the SQI. Our artifact models demonstrate that artifact types have differing effects on systolic, diastolic and mean ABP estimates. We show that, for most artifact types, diastolic ABP estimates are less noise-sensitive than mean ABP estimates, which in turn are more robust than systolic ABP estimates. We also show that our SQI can provide error bounds for both HR and ABP estimates. The KF/SQI-fusion method described in this article was shown to provide an accurate estimate of blood pressure and HR derived from the ABP waveform even in the presence of high levels of persistent noise and artifact, and during extreme bradycardia and tachycardia. Differences in error between artifact types, measurement sensors and the quality of the source signal can be factored into physiological estimation using an unbiased adaptive filter, signal innovation and signal quality measures.
Article
Motion artefact corruption of pulse oximeter output, causing both measurement inaccuracies and false alarm conditions, is a primary restriction in the current clinical practice and future applications of this useful technique. Artefact reduction in photoplethysmography (PPG), and therefore by application in pulse oximetry, is demonstrated using a novel non-linear methodology recently proposed by the authors. The significance of these processed PPG signals for pulse oximetry measurement is discussed, with particular attention to the normalization inherent in the artefact reduction process. Quantitative experimental investigation of the performance of PPG artefact reduction is then utilized to evaluate this technology for application to pulse oximetry. While the successfully demonstrated reduction of severe artefacts may widen the applicability of all PPG technologies and decrease the occurrence of pulse oximeter false alarms, the observed reduction of slight artefacts suggests that many such effects may go unnoticed in clinical practice. The signal processing and output averaging used in most commercial oximeters can incorporate these artefact errors into the output, while masking the true PPG signal corruption. It is therefore suggested that PPG artefact reduction should be incorporated into conventional pulse oximetry measurement, even in the absence of end-user artefact problems.
Conference Paper
Development and evaluation of automated decision support systems requires large amounts of well-characterized, reproducible test data. The MIMIC (Multi-parameter Intelligent Monitoring for Intensive Care) Database is intended to meet these needs. The database, currently nearing completion, will include 100 patient records, each typically containing between 24 and 48 hours of continuous data recorded from patient monitors in the medical, surgical, and cardiac intensive care units of Boston's Beth Israel Hospital. Each record will be accompanied by detailed clinical data derived from the patient's medical record and from the hospital's on-line medical information systems. We select patients to record from those likely to be hemodynamically unstable during the planned recording period. We expect to complete the selection of the recordings to be included in the database by the end of 1996, and to make the database available to other researchers shortly thereafter.
Article
In this paper, a methodology for using pulse photoplethysmography (PPG) signal to automatically detect sleep apnea is proposed. The hypothesis is that decreases in the amplitude fluctuations of PPG (DAP), are originated by discharges of the sympathetic branch of autonomic nervous system, related to arousals caused by apnea. To test this hypothesis, an automatic system to detect DAP events is proposed.The detector was evaluated using real signals, and tested on a clinical experiment. The overall data set used in the studies includes the polysomnographic records of 26 children which were further subdivided depending on the evaluation of interest. For real signals, the sensitivity and positive predictive value of the DAP detector were 76% and 73%, respectively. An apnea detector has been developed to analyze the relationship between apneas and DAP, indicating that DAP events increase by about 15% when an apnea occurs compared to when apneas do not occur. A clinical study evaluating the diagnostic power of DAP in sleep apnea in children was carried out. The DAP per hour ratio rDAP was statistically significant (p=0.033) in classifying children as either normal rDAP=13.5±6.35 (mean ± S.D.) or pathologic rDAP=21.1±8.93.These results indicate a correlation between apneic events and DAP events, which suggests that DAP events could provide relevant information in sleep studies. Therefore, PPG signals might be useful in the diagnosis of OSAS.
Further analysis and development of a method for quantification of EEG characteristics, proposed by Hjorth (1970), show that the proposed time domain descriptors convey all information which is obtainable from conventional frequency analysis. The application of the descriptors to superimposed responses of general physical systems reveals that they convey information which is highly relevant for describing the generating system. Thus, the descriptors not only constitute a tool for a simplified quantitative analysis of the curve characteristics, but they also offer interpretations in terms of physical models suggesting that the descriptors, or other parameters derived from the descriptors. may have physiological correlates. Such a generalized physical approach to the EEG is supported by clinical results which indicate that the descriptors convey basic information concerning physiological states.RésuméL'analyse plus poussée et le développement d'une méthode de quantification des caractéristiques EEG proposée par Hjorth (1970), montrent que les descripteurs de series temporelles proposés transmettent toute l'information qu'il est possible d'obtenir à partir de l'analyse de fréquences. L'application de ces descripteurs à des réponses surimposées de systèmes physiques généraux révèle qu'ils transmettent des informations hautement adéquates à la description du système générateur. Ainsi, ces descripteurs constituent non seulement un outil d'analyse quantitative simplifiée des caractéristiques des courbes, mais ils offrent également des interprétations en termes de modèles physiques suggérent que les descripteurs, ou d'autres paramètres dérivés des descripteurs, puissent avoir des corrélations physiologiques. Une telle approche physique généralisée à l'EEG est validée par les résultats cliniques qui indiquent que les descripteurs transmettent l'information de base concernant les stades physiologiques.