See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/301589275

Development of low cost wireless biosignal acquisition system for ECG EMG
and EOG

Conference Paper · December 2015

DOI: 10.1109/EICT.2015.7391945

CITATIONS READS

13 1,472

4 authors, including:

Md Asif Ahamed Md. Hanif Ali Sohag

Daffodil International University University of Saskatchewan
9 PUBLICATIONS   36 CITATIONS    15 PUBLICATIONS   43 CITATIONS   

SEE PROFILE SEE PROFILE

Mohiuddin Ahmad
Khulna University of Engineering and Technology
189 PUBLICATIONS   898 CITATIONS   

SEE PROFILE

Some of the authors of this publication are also working on these related projects:

CP # 3472, Dept. of BME, KUET View project

Wavelet Analysis Based Classification of Emotion from EEG Signal View project

All content following this page was uploaded by Md Asif Ahamed on 27 April 2016.

The user has requested enhancement of the downloaded file.

 Proceedings of International Conference on Electrical Information and Communication Technology (EICT 2015)

Note: This is the Pre-print Version

Development of Low Cost Wireless Biosignal

Acquisition System for ECG EMG and EOG

Md. Asif Ahamed1*, Md. Asraf-Ul- Ahad2, Md. Hanif Ali Sohag1, Mohiuddin Ahmad1
1
Department of Electrical and Electronic Engineering
2
Department of Conservative Dentistry and Endodontics
1
Khulna University of Engineering & Technology, Khulna-9203, Bangladesh
2
Bangabandhu Sheikh Mujib Medical University, Dhaka- 1214, Bangladesh
*E-mail: asif.fx@live.com

Abstract— A biosignal is a human body variable that can be application of EMG and EOG is human machine interface and
measured and monitored continuously and provide information controlling of various devices.
about the health status. Among them well known bioelectrical
signals are Electrocardiograph (ECG), Electromyography Our research aims to develop a low cost biosignal
(EMG), Electroencephalogram (EEG) and Electrooculogram acquisition system which is affordable for the people of
(EOG). Those signals are useful for different applications like developing and under developed country. In our previous
disease diagnosis, human machine interface, entertainment. This papers, we have developed an ECG data acquisition system and
paper presents a low cost, wireless biosignal acquisition system an ECG monitoring system [6,7]. In this research, we have
specialized for ECG, EMG and EOG. In this system, Arduino modified the system for acquisition of ECG, EOG and EMG
Uno in used and an application is developed for visualizing and together. This paper presents a prototype biosignal acquisition
storing the signal in real-time. This application is developed by system, which is portable, battery powered and it includes
processing. It stores the signal data in a text file, which can be wireless facility. In this system biosignal is transferred by
used in MATLAB for analysis. In this system, biosignal is Bluetooth serial communication and it gives wireless
transferred by Bluetooth serial communication, which ensures connectivity up to 9 meter. The equipment used in this system
safety and reduces noise interference. This system can be used consumes low power and this system is battery powered. For
either Windows, Linux, Mac OS and suitable for both laptop and this it can be used during load shedding and on those remote
desktop computer.
areas where electricity have not reached.
Keywords—Biosignal acquisition, ECG, EMG, EOG, Powerline This paper is organized as follows: Section II shows the
interference architecture of the system. Section III describes the proposed
methodology. Section IV explain the software implementation.
I. INTRODUCTION Experimental results are illustrated in section V. Finally,
Human body generates various different signals. Biosignal section VI concludes the results.
can be defined as any signal in living beings that can be
continually measured and monitored. It can be both electrical II. SYSTEM ARCHITECHURE
and non-electrical signals. Bioelectrical signals can be sensed This system consist of three electrodes. Figure 1 shows the
by non-invasive method by using skin-surface transducer,
which is commonly used for Electrocardiograph (ECG),
Electromyography (EMG), Electroencephalogram (EEG) and
electrooculogram (EOG) [1]. Both ECG and EOG signals are
bipolar low-frequency signal. The normal range of ECG signal
is 0.05-100Hz having its amplitude ranges from 10µV to 5mV,
whose typical value is 1mV [2]. EOG signal ranges between 15
μV to 200 μV with a frequency range of about 0-30 Hz [3].
Generally the amplitude of the EMG signal ranges from 0-
10mV (peak-to-peak) and its frequency ranges from 0- 500Hz
[4]. The dominant energy in EMG is being in 50-150Hz [4].
The frequency information of those signal is highly useful
for disease diagnosis. Among those signals ECG is very useful
for diagnosis especially for heart related diseases and disorders
such as, cardiovascular disease, pulmonary disease, sudden
cardiac arrest, etc [5]. EMG is used clinically for diagnosis of
neurological and neuromuscular problems. Another important Fig. 1. System Architecture of the Biosignal acquisition system.
 Saturation of amplifier generally occurs due to amplification of
electrode offset voltages [10]. To set the gain of AD620, a
resistor R is used and its value was 3kΩ. The gain expression
for AD 620 is given in Eq. (1).
. Ω
Gain 1 (1)
R

For removing the low frequency noises first a high pass

filter of 0.03 Hz cut off frequency is used after first stage
amplification. After second stage amplification, a band pass
filter is implemented using a high pass and a low pass filter in
cascade. For ECG, EOG and EMG cut off frequency of all high
(a) (b) pass filter is 0.03 Hz. For ECG cut off frequency of the low
pass filter is 159 Hz. For EMG it is 564 Hz and for EOG it is
15 Hz. Three switches S1, S2 and S3 are used for filtering path
selection of the ECG, EOG and EMG signals. All those filters
are passive First order filter. General expression for cut off
frequency of the filter is given in Eq. (2) [11].
F (2)
R C

(c)

Fig. 2. (a) limb leads and augmented limb leads for ECG, (b) Electrode
position for horiontal and vrtical eye mvemrnt, (c) Electrode position for
biceps EMG.

system architecture of this system. In this system, Arduino Uno

is used for analog to digital conversion and signal transmission.
By this system ECG can be taken from both limb leads and
augmented limb leads. Figure 2 shows the electrode position
for ECG, EOG and EMG.
III. PROPOSED METHODOLOGY
The proposed system consists of four units: sensor unit,
signal conditioning unit, power supply unit and Microcontroller
unit.
A. Sensor Unit Fig. 3. Circuit diagram of the system.
For sensing ECG, EMG and EOG signal, silver chloride
electrodes are used. A commercially available disposable C. Power Supply Unit
electrode named Bio Protech T716 was used, which is shown In this system two commercially available 9V disposable
in figure 4 (b) [8]. Lithium battery is used. A voltage regulator LM7805 is used
for safety and short circuit protection. It gives regulated output
B. Signal Conditioning Unit voltage of ±5V. In this system neutral electrode must connect
Signal conditioning unit consist of instrumentation to the 0V of the dual supply.
amplifier, op-amp and filter. In the top of figure 3 signal
conditioning circuit is shown. In this system, a commercially D. Microcontroller Unit
available low cost instrumentation amplifier AD620 is used, This unit consists Arduino Uno board and Bluetooth
which gives Common-mode rejection ratio (CMRR) greater module HC-06. Arduino UNO board has a 6 channel 10-bit
than 100dB at nearly 1 kHz [9]. Total gain of 840 is achieved analog to digital converter (ADC) and it returns a linear value
by two stage amplification. First stage amplification was done from 0 to 1023 corresponding to 0V and +5V respectively [12].
by AD620 and the gain was 13.7. Second stage amplification is As the ADC works between 0 to 5V, a 3.3V virtual ground is
done by CA3130 op-amp and the gain of second stage was 51. used for making the bipolar signal to unipolar. For serial
CA3130 was used as a non-inverting amplifier. To avoid the transmission baud rate is taken as 38400 bps and sampling rate
risk of amplifier saturation two stage amplification is done. of 1270 samples per second is obtained. A Bluetooth module
HC-06 does signal transmission between Arduino UNO and For visualizing and storing the signal a laptop running on
laptop. Bluetooth module HC-06 has a default baud rate of Windows 8.1 operating system having Intel B820 processor
9600 bps, which was changed to 38400 bps by a software and 4GB ram was used. ECG, EMG and EOG signals were
named putty. taken from a boy of 23 years old. The ECG signal was taken in
Lead I. In Figure 6, every red line indicates 1V of total 5V and
both ECG and EMG signals are above 3V. This is because in
this system 3.3V virtual ground is used. Stored ECG, EMG and
EOG signals in text file are imported in MATLAB for analysis.
MATLAB R2014a is used for analysis.

(a) (b)
Fig. 4. (a) Designed ±5V dual power supply. (b) Bio Protech T716 ECG
electrode..

IV. SOFTWARE IMPLEMENTATION (a) (b)

Arduino Uno board has a microcontroller named Fig. 6. Output of the application for (a) ECG and (b) Biceps EMG.
ATmega328. For programming the microcontroller, integrated
development environment provided by the Arduino platform is
used. For visualizing and storing the signal an application is
developed by Processing. Processing is an open source
programming language and integrated development
environment (IDE). It builds on the Java language, but uses a
simplified syntax and graphics programming model and
available for Linux, Mac and Windows operating system. For
receiving the signal data serial library of processing is used. In
processing line() function is used for visualizing the signal and Fig. 7. ECG signal from signal conditionig circuit.
println() function is used for sorting the signal, which creates a
txt file and saves the values received from Arduino ADC [13].
V. EXPERIMENTAL MEASUREMENTS AND RESULTS
Figure 5 shows the experimental measurement performed
by us for ECG signal and the prototype biosignal acquisition
system.

Fig. 8. Frequency spectrum of the ECG signal.

Figure 7 shows the ECG signal from signal conditioning

circuit output and this signal is contaminated with power line
interference (50Hz). This is because of common-mode voltage
produced by the ac mains of 50 Hz frequency with line voltage
220V [14].

Fig. 9. ECG signal after filtering in MATLAB.

Figure 8 shows the frequency spectrum of the contaminated

Fig. 5. Experimental measurements performed by us for ECG. ECG signal. In this figure, magnitude for 50Hz is 0.001738.
Figure 9 shows the filtered ECG signal. For filtering a TABLE I. COMPONENT LIST FOR THIS SYSTEM
Butterworth band pass and an elliptic band stop filter of 50Hz Component Name Quantity Unit Total Cost
was used. Lower and higher cut off frequency of Butterworth Price (USD)
filter was .033Hz and 150 Hz. (USD)
Arduino UNO 1 9 9
Bluetooth Dongle 1 2 2
Bluetooth Module HC-06 1 6 6
AD620 IA 1 6 6
Resistor and Capacitor 20 0.01 0.2
LM 7805 2 0.15 0.3
CA3130 op-amp 1 0.25 0.25
Electrodes 3 0.09 0.27
9V Battery 2 0.5 1
Fig. 10. EOG signal for horizontal eyemovement. Total Cost of the system ( USD) 25.02

Figure 10 shows the EOG signal for horizontal eye

movement. When the eye moves left, a positive peak is found VI. CONCLUSION
and when the eye moves right, a negative peak is found. In The main aim of this research has been fulfilled. We found
figure 11, the frequency of the EOG signal ranges from 0- the desired ECG, EMG and EOG signal and their frequencies
10Hz. As a low pass filter of 15 Hz is used, power line in a normal range. In this system we have used simple RC
interference (50 Hz) does not exist. filtering and we have not implemented any analog notch filter
of 50Hz. But as the signal conditioning circuit is battery
powered and the signal is transferred wirelessly, noise
interference in reduced and there exist a very small amount of
power line interference (50Hz). This is because of electric field
coupling from the nearby power line to the body. The proposed
hardware circuit is simple and effective for ECG, EMG and
EOG data acquisition. The cost of this system is only USD
25.02. The component used for this system consumes very low
Fig. 11. Frequency spectrum of the EOG signal. power. It takes maximum current of 60mA and it can function
up to 22 hours continuously by using two 9V battery.
In future, we will develop the robust biosignal acquisition
system with various control supports and protection schemes.

ACKNOWLEDGMENT
This work was partially supported by Higher Education
Quality Enhancement Project (HEQEP), UGC, under Sub-
project “Postgraduate Research in BME”, CP#3472,
Fig. 12. Biceps EMG signal. Bangladesh.

For Biceps EMG signal acquisition electrodes were placed REFERENCES

according to figure 2 (c). Figure 12 shows the biceps EMG [1] E. Kaniusas, “Biomedical signals and sensors I: Linking physiological
signal taken from left hand. As the force of the muscle phenomena and biosignals”, Springer Science & Business Media, 2012.
increases the peak of the EMG signal increases. In figure 13 [2] A. M. Al-Busaidi and L. Khriji, “Digitally Filtered ECG Signal Using
Low-Cost Microcontroller,” in International Conference on Control,
the frequency of the EMG signal ranges from 0-150Hz. There Decision and Information Technologies (CoDIT), pp. 258-263, 2013.
exists a small amount of power line interference same as the
[3] B. S. M Parameshwari, B. S. Madhurya, and K. U. Rao, “Preparation of
ECG signal and the magnitude of 50 Hz is found 0.0015. papers a cost effective prototype for electrooculogram for effective eye
tracking,” International Journal of Innovation, Management and
Technology , vol. 4, no. 5, pp.494-497, Ocotober 2013.
[4] H. Dong-Mei, Y. Yi, and W. Zheng. “Measurement System for Surface
Electromyogram and Handgrip Force Based on LabVIEW”, In World
Congress on Medical Physics and Biomedical Engineering, Springer
Berlin Heidelberg, pp. 67-70, 2009.
[5] M. K Islam, A. N. M. M Haque, G. Tangim, T. Ahammad, and M. R. H.
Khondokar, “Study and Analysis of ECG Signal Using MATLAB &
LABVIEW as Effective Tools, International Journal of Computer and
Electrical Engineering, Vol.4, No.3, pp. 404-408, 2012.
[6] M. A. Ahamed, M. A. Ahad, M. H. A. Sohag, and M. Ahmad.
Fig. 13. Frequency spectrum of the EMG signal. “Development of Low Cost Wireless ECG Data Acquisition System,” in
 International Conference on Advances in Electrical Engineering [10] D. Prutchi, M. Norris, “Design and Development of Medical Electronic
(ICAEE), 2015. Instrumentation: A Practical Perspective of the Design, Construction,
[7] M. A. Ahamed, M. K. Hasan, M. S. Alam, “Design and Implementation and Test of Medical Devices,” John Wiley & Sons, 2005.
of Low Cost ECG Monitoring System for the Patient Using [11] R. Raut and M. N. S. Swamy, “Modern Analog Filter Analysis and
Smartphone,” in International Conference on Electrical & Electronic Design: A Practical Approach,” John Wiley & Sons, 2011.
Engineering (ICEEE), 2015. [12] J. Oxer and H. Blemings, “Practical Arduino: Cool Projects for Open
[8] Bio Protech ECG Electrode specification [Online]. Available: Source Hardware,” Apress, 2009.
http://sanjayintl.com/pdf/ecg_bioprotech.pdf [13] C. Reas and B. Fry,“Processing: a programming handbook for visual
[9] Datasheet of AD 620 instrumentation amplifier [Online]. Available: designers and artists”, Mit Press, 2007.
http://www.analog.com/media/en/technical-documentation/data- [14] A. Brahme,“Comprehensive Biomedical Physics”, Newnes, 2014.
sheets/AD620.pdf

View publication stats