ELECTROCARDIOGRAM RHYTHM SIMULATION IN OPEN

SOURCE ENVIRONMENT

A Contribution to Training in Biomedical Sciences

A. Cardoso Martins, P. Dias Costa, J. Miguel Marques and R. Cruz Correia

Department of Biostatistics and Medical Informatics, Faculty of Medicine, University of Porto, Portugal

Keywords: Electrocardiogram, ECG, Training, Software, Open source, Computer simulation.

Abstract: Considering the importance of training and the need to identify some important electrocardiographic

patterns, our aim was to develop a software-based electrocardiogram rhythm simulator, able to generate

signals for purposes of training health care professionals, freely available. Additionally we intended to

implement a training assessment and the ability to configure specific rhythms/patterns with user-defined

settings.

In order to achieve the availability requirement we decided to develop our tool as open source. All the

software was developed in Linux platform, but releases are available in binary format for Linux and

Microsoft Windows platform.

The result was a functional signal simulator which satisfied all the requirements we proposed in our aim.

The final result of our work is presented and available at Sourceforge.

1 INTRODUCTION

Electrocardiographic signal is a graphic

representation of the electrical activation

(depolarization) and repolarization of cardiac cells,

as a result of the expression of main activation

vector in each lead. In an electrocardiogram (ECG)

strip it is possible to identify several incidents, that

concern specific electrical activity. Additionally it is

possible to define a number of intervals and

segments between these main incidents, that concern

stimuli conduction and that do have clinical

correspondence, with a determined “normal”

duration, amplitude and morphology (Lipman and

Cascio, 2001; Wagner, 2006).

Presently, and more than 100 years passed over its

introduction in clinical practice by Willem

Einthoven (Moukabary , 2007), the ECG is still a

valuable diagnostic instrument. The astonishing

technological evolution felt in the medical field in

the past 50 years, with particular relevance in

Cardiology, created the misconception that newer

techniques have replaced electrocardiography. Not

only is the electrocardiogram an important

diagnostic tool, but also to guide the treatment of

common severe cardiac conditions, and determine

patients prognosis(Hurst, 2003).

Training of biomedical professionals is crucial for

patient care. The ability to train procedures, establish

diagnosis and define prognosis of a disease prior to

applying that knowledge in real-life patients is an

enormous advantage for both professionals and

patients, with clinical, legal, and ethical

implications. In terms of ECG, the wide variety of

normal patterns, sometimes quite different from each

other, and the difficulty to identify some subtle

abnormal patterns makes prior training not only

necessary but also imperative. The

acknowledgement of training needs, lead to the

development of a joint guideline for training in

electrocardiography (Kadish et al., 2001).

A wide variety of ECG simulators, electronic or

software-based, both commercial and non-

commercial, are currently available. A series of

commercial devices, using just an electronic

approach, were built especially for ECG rhythm

simulation and training. Nowadays they are

currently in use, but only for specific training and

maintenance of medical equipment. With the

explosion of computer technology back in the

1970's and 1980's several computer programs

6,7

were

developed for ECG and pacing rhythm training and

512

Cardoso Martins A., Dias Costa P., Miguel Marques J. and Cruz Correia R. (2010).

ELECTROCARDIOGRAM RHYTHM SIMULATION IN OPEN SOURCE ENVIRONMENT - A Contribution to Training in Biomedical Sciences.

In Proceedings of the Third International Conference on Health Informatics, pages 512-516

DOI: 10.5220/0002766805120516

 SciTePress

simulation. Those systems seem to have been

abandoned (at least for what concerns free software

implementations), with a few exceptions, probably

due to severe limitations namely in scientific

contents. Conversely, a series of well developed

ECG rhythm simulation programs is currently

available for commercial use. Additionally, an

unaccounted number of advanced life support (ALS)

manikins are also available, allowing the chance to

diagnosis and treat malignant arrhythmias (Mueller

et al., 2005; Schlindwein et al., 2002; Gordon,

1974). However these training manikins are

somewhat limited in their ability to deliver and

simulate certain electrocardiographic patterns as

they are more emergency training oriented.

As far as we know, there is presently no ECG

simulation tool widely available to everyone, able to

receive suggestions and improvements from a wide

variety of professionals of medical and other

scientific areas, that can be freely improved and

reviewed by experts. As stated before, some

exceptions (Karthik, 2002) are available, but with

limitations, like the requirement of a third party

software license (Matlab) or lacking usability for

everyday usage.

Considering the importance of training and the need

to identify some important electrocardiographic

patterns, our aim was to develop a software-based

ECG rhythm simulator, able to generate ECG

signals to train health care professionals, freely

available. Additionally we intended to implement a

training assessment and the ability to configure

specific rhythms and patterns with customizable

settings. In order to achieve the availability

requirement we decided to develop our tool as open

source, therefore we have chosen the GNU Public

License (GNU General Public License, 2009) GPL

v3 license model.

2 METHODS

We performed a search in the Pubmed (PubMed,

2009) reference database regardless of the type of

publication. For addressing the database, we used

the query: “electrocardiogram [MeSH] AND

training [MeSH] AND simulator” with limits

selected for “English language” and “Humans”. The

query was performed on the 15

May 2009 and

returned 6 articles for analysis. The analysis was

performed by all investigators. Selection criteria

were defined prior to the query and included: 1) type

of study; 2) critical analysis of the results, and 3)

suitability to our subject. After applying the above

criteria, we selected two articles (Fukushima e tal.,

1984; Gordon, 1974). This method, despite its

scientific reliability, does not take into account the

existence of other not so formal initiatives. To

understand the state of development of commercial

and non-commercial ECG simulation tools we

performed a qualitative survey using the same terms

on Google search engine. The result, with the

limitations of being a qualitative survey, made clear

that a wide variety of tools is available but not with

the type of development we tried to implement.

2.1 Implementation

As the first task of the project, we tried to find a

simple and flexible way to generate signals that

could be similar to those of human ECG with

different patterns. Fourier series were selected for

representing ECG signals, as any periodic functions

which satisfy Dirichlet's condition can be expressed

as a series of scaled magnitudes of sin and cos terms

of frequencies which occur as a multiple of

fundamental frequency. ECG signal is periodic with

fundamental frequency determined by the heat beat

(Karthik, 2002). Each significant feature of ECG

signal can be represented by shifted and scaled

versions of waveforms.

Figure 1: simECG version 1.0 screenshot showing custom

settings.

2.2 Environment

The Qt 4 development framework was chosen to

implement the project, as it is very flexible, uses

C++ as the programming language with easy-to-use

graphical user interface support, and is available in

many different operating system platforms. Another

ELECTROCARDIOGRAM RHYTHM SIMULATION IN OPEN SOURCE ENVIRONMENT - A Contribution to

Training in Biomedical Sciences

513

Figure 2: simECG activity (downloads and traffic) on Sourceforge.

reason, was that Qt (Qt4 cross-platform application

and user interface framework, 2009) version 4 has a

licensing model that allows the use of GPL to

develop an open source software solution. The

software was developed using many tools, however

the most important ones were Qt Creator (Qt Creator

Integrated Development Environment, 2009) as the

integrated development environment (IDE), and

Sourceforge (Sourceforge, 2009) as the project

management system that provided a complete

software build platform, that includes .svn for source

code repository, package release file system

repository, mailing lists and forum for developer

discussion, bug and feature tracking system, and a

project web page. All these resources are publicly

available (simECG sourceforge project page, ).

2.3 Programming

The architecture of the project lead to C++ classes

that implemented the intended block behavior.

“Signal Plotter” displays an ECG signal in a typical

ECG strip format, which was in turn created by

“ECG generator”, that calculates the signal based on

the above described Fourier series, adjusted by the

user settings. In turn, every setting, preset or custom,

is stored as an “ECG Presets” class. Every time the

user runs an assessment, its corresponding class is

instantiated. The assessment controls a set of 10

questions which the user should select the correct

one displayed. Three presets are randomly chosen,

but only one is accordingly displayed. The

myMainWindow class, implements the main

program window, menu toolbar, docking toolbar

with a tab separator, and all the graphical widgets

present. It is the class that glues all the code, mainly

through the concept of signals and slots, that is very

well known in Qt development.

2.4 Dissemination

The simECG application was disseminated via the

Sourceforge, with the first beta released on the 29

of May 2009. The simECG project included bugs,

features, and users/developers forum was created to

help disseminating the application. In the next

section we report the tracker activity since the first

release.

3 RESULTS

3.1 Application Features

simECG has several features, that allows the user to

display predefined ECG signals such as: sinus

rhythm, bradycardia or tachycardia; junctional

rhythm; accelerated junctional rhythm;

idioventricular rhythm; accelerated idioventricular

rhythm; supraventricular tachycardia; monomorphic

ventricular tachycardia (VT); ventricular fibrillation;

first degree atrio-ventricular (AV) block; and AV

dissociation. Additionally the user can also define

some custom settings, such as heart rate; PR

interval; P, QRS and T waves duration, amplitude

and polarity. Nevertheless, some bugs still persist

and are being corrected specially through the

Sourceforge community infrastructure.

The activity in the project, despite being considered

a recent release, seems to reflect the good adherence

from the community.

4 DISCUSSION AND

CONCLUSIONS

The simECG is able, as stated in our aim, to

HEALTHINF 2010 - International Conference on Health Informatics

514

generate ECG signals, for purposes of training

health care professionals as an open source solution,

freely available to the community from now on.

The final result of our work is presented and

available at Sourceforge (Sourceforge, 2009) in the

simECG project page (simECG sourceforge project

page, ) as version 1.0. All the software was

developed in Linux platform, but releases are

available in binary format for Linux and Microsoft

Windows platform.

With the release of simECG 1.0 the first steps

towards the building of a community were taken,

based on the premises that: all the components of the

project should be developed on Open Source

applications; all the contribution should be “non-

paid”; all the releases should be free; and the source

code should be available for everybody to read,

interpret and re-write. All those premises along with

the idea that the authors alone could not support

such a project, lead to the creation and maintenance

of simECG at Sourceforge.

With just a few months online, the authors think the

project as a good chance of succeed. The download

and traffic rates are quite reasonable. Feedback from

users and participants has been good, although until

now we could only recruit one new contributor.

Nevertheless the project will audited at six months

in order to compare the web traffic with other

projects, namely in healthcare related projects, and

potential contributors. This is allow the authors to

identify more accurately the directions the project

should take.

4.1 Limitations and Future

Perspectives

At the moment simECG cannot be integrated in

other solutions, as it was designed to be a standalone

interface and does not provide “standards-based”

interface.

Authors also recognize that more experimental

results are needed, and a comparative evaluation

should be performed.

Future work includes implementation of some

features and custom settings, such as premature

atrial or ventricular contractions, atrial fibrillation

and some special conditions (second degree AV

blocks and polymorphic VT) that were implemented

in this version due to time constraints. For that same

reason, was not possible to implement the atrial

fibrillation pattern. The implementation of ECG

motion (rolling) needs further advanced research. In

future versions the authors intend to provide

additional features in the training presets and include

additional settings.

All pending bugs and requested features are

described in detail in the Sourceforge tracker

service, and need to be object of close attention in

future versions.

simECG will always be an unfinished project. In the

future authors intend to build and support a

community, with the intention of releasing new

improved versions.

ACKNOWLEDGEMENTS

The authors wish to thank R. Karthik, Sandra Jardim

and Ernesto Aranda for their help in the

development of this project.

REFERENCES

Lipman B, Cascio T, editors. ECG - avaliação e

interpretação. Loures: Lusociência; 2001. p. 1-22.

Wagner GS, editor. Marriott's Practical

Electrocardiography. 10th ed. Philadelphia: Lippincott

- Raven; 2006. p. 1-69.

Moukabary T.Willem Einthoven (1860-1927): Father of

electrocardiography. Cardiol J 2007;14(3):316-7.

Hurst JW. Current status of clinical electrocardiography

with suggestions for the improvement of the

interpretive process. Am J Cardiol 2003; 92(9):1072-9.

Kadish AH, Buxton AE, Kennedy HL, Knight BP, Mason

JW, Schuger CD, et al. ACC/AHA clinical competence

statement on electrocardiography and ambulatory

electrocardiography: a report of the American College

of Cardiology/American Heart Association/American

College of Physicians-American Society of Internal

Medicine Task Force on Clinical Competence

(ACC/AHA Committee to Develop a Clinical

Competence Statement on Electrocardiography and

Ambulatory Electrocardiography). Circulation 2001;

104: 3169-78.

Fukushima M, Inoue M, Fukunami M, Ishikawa K, Inada

H, Abe H. Computer-assisted education system for

arrhythmia (CAESAR). Comput Biomed Res 1984;

17(4):376-88.

Battersby EJ. Computer assisted instruction in the

diagnosis of cardiac arrhythmias. In: Fuller HW, Ward

JE, Bonn JE, Clippinger RF, Burrows JH, Connelly

ME, et al, editors. American Federation of Information

Processing Societies - Spring Joint Computer

Conference; 1969 May 14-16; Boston, Massachusetts.

New Jersey: AFIPS Press; p.703-8.

Mueller MP, Christ T, Dobrev D, Nitsche I, Stehr SN,

Ravens U, Koch T. Teaching antiarrhythmic therapy

and ECG in simulator-based interdisciplinary

ELECTROCARDIOGRAM RHYTHM SIMULATION IN OPEN SOURCE ENVIRONMENT - A Contribution to

Training in Biomedical Sciences

515

undergraduate medical education. Br J Anaesth 2005;

95(3):300-4.

Schlindwein M, von Wagner G, Kirst M, Rajewicz M, Karl

F, Schöchlin J, Bolz A. Mobile patient simulator for

resuscitation training with automatic external

defibrillators. Biomed Tech 2002;47 (Suppl 1):559-60.

Gordon MS. Cardiology patient simulator. Development

of an animated manikin to teach cardiovascular

disease. Am J Cardiol 1974; 34(3):350-5.

Karthik R. ECG Simulation using Matlab – Principle of

Fourier Series. Anna University, 2002.

PubMed [serial online]. 2009 [cited 2009 May 15; 1(1):48

screens]. Accessible at: http://www.pubmed.gov.

GNU General Public License [serial online]. 2009.

Accessible at: http://www.gnu.org.

Qt4 cross-platform application and user interface

framework [serial online]. 2009. Accessible at:

http://www.qtsoftware.com/products.

Qt Creator Integrated Development Environment [serial

online]. 2009. Accessible at:

http://www.qtsoftware.com/products/developer-tools.

Sourceforge [serial online]. 2009. Accessible at:

http://sourceforge.net.

simECG sourceforge project page [serial online].

Accessible at: http://sourceforge.net/projects/simecg

HEALTHINF 2010 - International Conference on Health Informatics

516