=Paper=
{{Paper
|id=Vol-2009/fmt-proceedings-2017-paper5
|storemode=property
|title=LifeStream: Prototype Implementation of Monitoring System for Dispatch Life Support
|pdfUrl=https://ceur-ws.org/Vol-2009/fmt-proceedings-2017-paper5.pdf
|volume=Vol-2009
|authors=Florian Grassinger,Jakob Doppler,Markus Wagner,Wolfgang Aigner
|dblpUrl=https://dblp.org/rec/conf/fmt/GrassingerD0A17
}}
==LifeStream: Prototype Implementation of Monitoring System for Dispatch Life Support==
https://ceur-ws.org/Vol-2009/fmt-proceedings-2017-paper5.pdf
LifeStream: Design and Prototypical Implementation
of a Monitoring System for Dispatch Life Support
Florian Grassinger, Jakob Doppler, Markus Wagner and Wolfgang Aigner
Institute of Creative\Media/Technologies, St. Pölten University of Applied Sciences, Austria
first.lastname@fhstp.ac.at
Abstract—Most laypersons who reanimate for the first time do detection and compares them to existing standards and there-
it inappropriately. Until now the only way to review the ongoing fore enhance the overall reanimation process for the dispatcher
reanimation was verbal feedback by the dispatcher on the phone, and the layperson.
who has only limited resources in order to review the reanimation
process. To overcome this issue, we designed and implmemented The major contribution of this paper is a functional prototype
LifeStream, a system using current smartphone technologies in that was tested during user tests and evaluated as well as a
order to measure reanimation parameters: chest compression straightforward experimental implementation.
rate (CCR) and chest compression depth (CCD). The system
is based on a server, web client and mobile application, which II. R ELATED WORK
gathers, processes and transfers the data. The development of
algorithms for CCR and CCD detection as well as the evaluation There are a number of tools and research work, that deal
of the system functionality is part of this paper. We conducted with the quality of CPR and its enhancement. However, none
a 2-day user test, where we compared the guided standard of them transmits the data in real time to an emergency medi-
reanimation process to the application supported process. The cal dispatcher (EMD). PocketCPR 1 is a mechanical device that
results of the tests showed that it is possible to develop an enhances the quality of CPR by simple audiovisual feedback in
application, which runs for at least ten minutes (crucial time till
ambulance arrives) and enhances the whole reanimation cycle real time, which was already evaluated [6]. The mobile version
for laypersons and dispatchers [1]. is called ZOLL PocketCPR, 2 which gives real time feedback of
an ongoing CPR through the smartphone. It uses smartphone
I. I NTRODUCTION sensors to give the user audio-visual feedback and introduces
Our fast aging population results in an increase of out-of- the user to the whole process of CPR. CPREzy 3 , is designed
hospital cardiac arrest situations. Often dispatch life support for CPR assistance and offers a simple interaction. It has an
and Cardiopulmonary Resuscitation (CPR) interventions are audible chirp and visual light pacing system with a metronome
performed by untrained laypersons and bystanders rather than to guide the CPR. In a study the device was compared with a
medical professionals [2]. The fear of making bad decisions normal reanimation and the results have shown, that there was
often restrains people from helping and saving life’s or bridge no significant difference in compression rate or duty cycles
the critical minutes until the ambulance arrives [3]. Time between the techniques [7]. Song et al. [8] describe the usage
critical medical emergency situations are situations where a of an inbuilt accelerometer sensor in smartphones to enhance
proper execution of all steps in the chain of survival is crucial the quality of a CPR by directly measuring the CCR and CCD.
and therefore every second counts [4]. A CPR often requires The main difference is that the feedback is restricted to the
immediate reaction and even if the chest compressions are not user and not an EMD. Up to now, like the ones mentioned,
totally appropriate, the attempt is crucial to save a person’s life. have begun to examine how to enhance the quality of CPR.
Over the past years, cardiopulmonary resuscitation has contin- But none of these studies concentrate on direct user feedback
uously improved and was further investigated by Roessler et and on the dispatch of crucial CPR parameters to the EMD.
al. [5].
Today’s smartphones are equipped with multimodal sensors III. BACKGROUND
to measure important context and even vital parameters that Using an accelerometer the following physical and technical
can be used to assess the situation during a reanimation. backgrounds should be considered:
For the development of a functional prototype, which assists
A. Physical considerations of spatio-temporal parameters
laypersons or unexperienced people by performing chest com-
pressions, we used the accelerometer sensor of the smartphone. The algorithms for CCR and CCD detection are based
Additionally, we utilized the network connectivity, as well as around the physical concept of acceleration and its first and
maintain an ongoing phone call and perform background tasks second order integral velocity and distance. Any change in the
such as transmitting real-time data to develop an effective velocity of an object results in acceleration. So acceleration
algorithm for chest compression rate and depth detection. 1 http://www.zoll.com/de/produkte/pocketcpr/, Accessed: April 11, 2016
The main goal of this research is to present a prototypical 2 https://goo.gl/B1pdMR, Accessed: April 11, 2016
implementation of a system which uses algorithms for chest 3 http://www.heartworkscpr.com/cprezy-facts.html, Accessed: April 16,
compression rate (CCR) and chest compression depth (CCD) 2016
41
LifeStream: Prototype Implementation of Monitoring System for Dispatch Life Support
Fig. 1. The website client and the application. On the left side is the starting point of the application and on the right side the results view after the application
was finished (mainly for testing purpose and recording).
is related to velocity, or depends on the change of it [9].
The relation of acceleration, displacement and velocity is
important, as all of these three quantities are vector quantities
(give information about direction) [10]. Using the example of
displacement (the directed distance between two points A and
B), it is theoretically possible to determine the final position
of the mobile phone, if it is used for CCD detection. The
distance would be wrong, as it’s only a scalar and counts up
the traveled way and not the direct way between two points.
Frequency detection is also possible, because after a certain
push threshold is exceeded, the push is correct and this counts Fig. 2. Reanimation with phone and LifeStream-App.
to the total frequency.
IV. D ESIGN & I MPLEMENTATION OF L IFE S TREAM
B. Technical considerations
Based on the input of project members and partners the
The accelerometer is a powerful mechanical low cost sensor, requirements for the main prototype were formulated: A
which is implemented into nearly every smartphone. It offers mobile client with a medical dispatch visualization server to
the possibility to measure the acceleration in a specified handle clients and a visualization website for visualizing data.
direction. The values measured by the smartphone are in m/s2
and always include the acceleration and deacceleration [11].
A. Usage scenario
In essence the accelerometer measures force that is applied not
acceleration. Acceleration just causes an inertial force that is When receiving an emergency call, the EMD advises the
captured by the force detection mechanism of the accelerom- caller to open the application (if not open already). Then the
eter or acceleration is the amount of force needed to move application registers at the server endpoint of the medical
each unit of mass. All calculations take place directly on the dispatch center and starts the streaming session. The EMD
smartphone and are processed further to the server, redirecting then instructs over the phone and guides the reanimating
them to the lifestream website (see 1 for visualization. For the layperson through the process. The phone has to be placed
prototype the visualization is restricted to one mobile client. between the hands and the victim. Although, many people
Later, each EMD has his own implementation in the already hesitate (results of user studies) to push directly on the phone,
established call taking system where it shows the visualization it won’t crack in most cases as the hands are laying flat on
of an ongoing CPR. the phone. Figure 2 shows the placement of the hands.
42
LifeStream: Prototype Implementation of Monitoring System for Dispatch Life Support
During the usage scenario definition and the continuous C. Calculation restriction:
collaboration with the project partners and members, the Physically and theoretically it should be possible to cal-
following four main design considerations were defined: culate the traveled distance of the phone by using the ac-
1) Simple usability: The application must be simple and celerometer. If the acceleration is integrated once, the result
has to gather data, perform calculations, transmit it and is the velocity of the object (in this case the smartphone).
stop the whole data acquisition and transmission process. After a second integration the result is the traveled distance
2) Restricted functionality: Frantic laypersons require an [9], [13]. Despite these equations seem fairly straightforward
application that is protected against unwanted termina- to implement, they are practically not possible. The natural
tion. This means the buttons, which are normally used spread error propagates problematically after each integration
to terminate the application or go back, were disabled. as well as the included gravitational force that applies to the
The application is also running in full screen mode and phone. A solution to this problem is the usage of a linear
it stays in wakeup state the whole time (10 minutes accelerometer, a sensor fusion of various other sensors that
minimum till ambulance arrives [1]). factors out the gravitational force. The main issue with using
3) Easy configuration: A simple and non-intrusive menu the above mentioned method is that accelerometers are bad
is used, which allows the change of parameters for the at dead-reckoning (continuous position determination). Ac-
calculation and termination of streaming. celerometers have some noise which varies from smartphone
4) Fast transmission: As a stable network connection to smartphone as each has its own manufacturer and device
cannot be granted the transmission has to be optimized. type. The noise can be filtered using various filter types, but
Therefore a small and simple data format (JSON [12]) normal accelerometers produce raw data, which is not filtered
is used, which already contains calculations. or smoothed. This noise will usually result in a non-zero mean,
that is continuously added and accumulates in the resulting
B. Server, Website & Mobile Client
velocity signal and later of course in the distance integration.
Server: The server is a basic NodeJS server which serves This behavior is called sensor drift, as the integration starts
the website and redirects the mobile clients. It allows bidirec- fairly well, but quickly accumulates the errors and the resulting
tional communication, so real-time communication is possible. values drift away.
The server distinguishes between normal clients (web) and Using the linear accelerometer of the Android system leads
mobile clients (Android), who transmit data to it. to better results, as the gravity is already removed and the re-
Website: The website combines various web technologies. sulting values are much smoother. After the gravity is removed
D3.js 4 is used for visualizing the data in a running line graph and the values are read and filtered with a respective filter, it is
in real time. The website can be reached over the domain advised to calculate the magnitude of the acceleration values
lifestream.fhstp.ac.at. At the current prototype state every web before continuing with further calculations [11].
client receives the website and while an Android client is
connected and streaming data, he can view the reanimation D. Calculation Solution:
data (seen in Figure 1). On the website the visualization is By taking all the previous problems and considerations into
separated into two major parts. First, the dynamically updating account, a final functional prototype was developed. The algo-
line chart that constantly plots the acquired reanimation data rithm is a very basic but powerful peak detection and frequency
from the smartphone. The color codes used for the frequency, estimator. After 15 seconds (an adequate update time, based
or pushes per minute, are abstractions based on the frequency on expert feedback) the frequency on the website is updated
range: based on the average reanimation frequency during this time.
• Red indicates a very bad frequency (all under 90 or above The frequency is calculated for these 15 seconds or any other
130). interval, approximated to one minute and then transmitted to
• Yellow indicates an average frequency (from 90 to 100 the server along with other values (e.g. approximate pressure
& 120 to 130 pushes). depth).
• Green indicates an optimal frequency (from 100 to 120 The optimal frequency of 100 pushes per minute should
pushes). theoretically be achieved by pushing always at least five
Second, the information section includes information about centimeters into the chest of the victim [2]. As CCD de-
the registered clients and the reanimation parameters, e.g., the tection with the given sensor and technology is not really
frequency. possible, the approach with frequency seemed more promising
Mobile Client: The mobile client is available for Android as well as an approximation of the distance based on the z-
devices and opens a stream to the server and transmits data of axis acceleration. As the performed reanimation of the user
an ongoing reanimation. The data acquisition, calculation and normally changes over time, especially when the power ceases,
transmission can be started with a simple button press, while the frequency detection is very difficult. The requirement to
the stop functionality is hidden in a small menu above along the algorithm must be to detect hard pushes as well as faint
other configuration options. pushes. Therefore, peak detection is implemented. According
to previous studies and extensive acceleration data logging
4 https://d3js.org/, Accessed: April 28, 2016 and plotting, the following concept was devised. Once the
43
LifeStream: Prototype Implementation of Monitoring System for Dispatch Life Support
• Displacement can be detected over a short amount of time
(movement along the z-axis).
• Displacement detection is not possible with the low cost
accelerometer.
VI. C ONCLUSION & L IMITATIONS
The performed tests have shown that available smartphone
accelerometers along with their embedding systems vary
widely and often heavily rely on the hardware and the algo-
rithm used. The accelerometer sensor is often erroneous and
creates a non-zero mean that adds up to further calculations.
The only solution is filtering and using a linear accelerometer.
Often enough the sensor samples slightly slower than the
actual sampling frequency as other tasks are more important
for the operating system in the background. That means for any
calculation it is problematic to rely on fixed time intervals as
they are often slightly shorter or longer. The errors are adding
Fig. 3. Peak detection and minimum threshold (along z-axis).
up over time and contribute heavily to the whole calculation. A
possible solution was to wait an offset time before calculating
acceleration values or signal traverse the zero line, a change and estimating depth. At least 100 ms proved to be useful
in acceleration happens and a peak can be detected (fig. 3). (discarding all before). A remaining problem is that some time
Thus, no matter how weak or hard the user pushes, the peak stamps are 100 ms or longer and also the fact that numerous
can be detected by its zero-line crossing and change of accel- important values are lost during the defined pause. During
eration with a minimal threshold of applied acceleration. The peak detection this can be fatal, as a global maximum could
algorithm is still based on the basic equation and calculation be skipped.
of the magnitude, During development it turned out that frequency detection is
much easier than continuous position determination, especially
V. E VALUATION & VALIDATION OF PROTOTYPE in smaller unit ranges (like cm). The theoretic (and physically
The algorithm was evaluated during a two-day user study, correct) equations are not feasible for usage in real world
which involved 25 laypersons between 18-50 years (14 lay applications. Accelerometers measure the acceleration in a
rescuers and eleven paramedic experienced). They have been body-fixed reference frame, where normally displacement in
selected randomly during an open experiment at St. Pölten earth-fixed reference frames is necessary. Therefore, it is not
University of Applied Sciences, Austria. Each volunteer was possible to only integrate the accelerometer twice and find
instructed before by two professional paramedics. The setup the displacement, except it is rotated into the earth fixed
included a reanimation phantom as well as a professional frame before the integration takes place. The project showed,
EMD, which was in his actual workplace. The participants that with the given premises of only using the low cost
were filmed during the whole process and the reanimation accelerometer sensor in smartphones, it is not possible to
phantom also recorded the reanimation process for later com- make a sturdy point about the displacement. Nevertheless,
parison to the algorithm. Each participant was not further it is possible to make a point about the current reanimation
instructed in the CPR process and they had to reanimate frequency very well by using the developed peak detection
(guided) for full ten minutes. Further randomization happened algorithm. Even a position determination could be possible
as some of the laypersons were just reanimating on the by using the peak detection and the currently viewed values
phantom without the mobile application. For further insight during the peak detection (a so-called window of values) for
in the evaluation and test scenario refer to [14] and [15]. the integration. As the values are always restricted to a certain
amount and interval, a double integration of those values
A. Results
would contain less errors that could add up over time.
The outcome of the tests clarified that a guided CPR by
using the system is far more efficient for both sides, the EMD ACKNOWLEDGMENTS
and the layperson, rather than a standard phone guided CPR. The authors would like to thank Stefan Loitzl and Peter
Some other interesting outcomes as well are: Pavlecka who have been former project members, contributed
• Most people hesitate to push on a phone, as it could crack. during development and lead the testing part. We also would
• The application detects the frequency very well and is like to give a special thanks to our project partners from Nortuf
comparable to a professional reanimation phantom that Niederösterreich especially Heinz Novosad and Raphael Van
is used for training purposes. However, the accuracy Tuldar for their support. This work was supported by the
is not as high as a professional sensor, compared to a Austrian Science Fund (FWF) via the “VisOnFire” project
smartphone accelerometer. (P27975-NBL).
44
LifeStream: Prototype Implementation of Monitoring System for Dispatch Life Support
R EFERENCES
[1] M. Ljunggren, M. Castrn, and K. Bohm, “Abstract 231: The Effect of
Dispatcher Call Processing Interval and Ambulance Response Interval
for Survival From Cardiac Arrest,” Circulation, vol. 128, no. Suppl 22,
pp. A231–A231, Jan. 2016.
[2] R. O. Cummins, J. P. Ornato, W. H. Thies, and P. E. Pepe, “Improving
survival from sudden cardiac arrest: the ”chain of survival” concept. a
statement for health professionals from the advanced cardiac life support
subcommittee and the emergency cardiac care committee, american heart
association.” Circulation, vol. 83, no. 5, pp. 1832–1847, 1991.
[3] L. Wik, J. Kramer-Johansen, H. Myklebust, H. Sreb, L. Svensson,
B. Fellows, and P. A. Steen, “Quality of cardiopulmonary resuscitation
during out-of-hospital cardiac arrest,” JAMA, vol. 293, no. 3, pp. 299–
304, Jan. 2005.
[4] C. Ro, “The ”chain of survival” concept: how it can save lives.” Heart
disease and stroke : a journal for primary care physicians, vol. 1, no. 1,
pp. 43–45, 1991.
[5] B. Roessler, R. Fleischhackl, H. Losert, C. Wandaller, J. Arrich,
M. Mittlboeck, H. Domanovits, and K. Hoerauf, “Cardiopulmonary
resuscitation and the 2005 universal algorithm: Has the quality of CPR
improved?” Wiener klinische Wochenschrift, vol. 121, no. 1-2, pp. 41–
46, Jan. 2009.
[6] S. Kirkbright, J. Finn, H. Tohira, A. Bremner, I. Jacobs, and A. Celenza,
“Audiovisual feedback device use by health care professionals during
CPR: A systematic review and meta-analysis of randomised and non-
randomised trials,” Resuscitation, vol. 85, no. 4, pp. 460–471, Apr. 2014.
[7] G. D. Perkins, C. Augr, H. Rogers, M. Allan, and D. R. Thickett,
“CPREzy: an evaluation during simulated cardiac arrest on a hospital
bed,” Resuscitation, vol. 64, no. 1, pp. 103–108.
[8] Y. Song, Y. Chee, J. Oh, and S. Lee, “Development of android
based chest compression feedback application using the accelerometer
in smartphone,” in Proceedings of the International Conference on
Biomedical Engineering and Systems, pp. 130–1.
[9] M. Mounts, “Physics hypertextbook,” vol. 44, pp. 1946–1947, 01 1970.
[10] J. Rybach, Physik fr Bachelors: mit 92 durchgerechneten Beispielen, 176
Testfragen mit Antworten sowie 93 bungsaufgaben mit kommentierten
Musterlsungen, 2nd ed. Mnchen: Fachbuchverl. Leipzig im Carl-
Hanser-Verl, 2010.
[11] G. Milette and A. Stroud, Professional Android sensor programming,
ser. Wrox Programmer to programmer. Indianapolis, Ind: Wiley, 2012.
[12] N. Nurseitov, M. Paulson, R. Reynolds, and C. Izurieta, “Comparison
of JSON and XML data interchange formats: a case study.” Caine, vol.
2009, pp. 157–162, 2009.
[13] P. Becker, “Erfassung und Verarbeitung von Sensordaten,” Hochschule
Bonn-Rhein-Sieg, 2013.
[14] P. Pavelcka, “LiFeStream - Validation of a mobile application for visual
assessment of cardiopulmonary resuscitation (CPR) acceleration data
compared with a life-support manikin sensory,” Masterthesis, FH St.
Pölten, St. Pölten, 2016.
[15] S. Loitzl, “Validating a sensor-based mobile health service with real-time
data acquisition for improving MPDS communication for lay-rescuer in
out-of-hospital cardiac arrest based on a randomized simulation trial,”
Masterthesis, FH St. Pölten, St. Pölten, 2016.
45