Progress In Electromagnetics Research Symposium Proceedings, KL, MALAYSIA, March 27–30, 2012 1901

Standalone ECG Monitoring System Using Digital Signal ProcessingHardware

Goh Chun Seng, Sh-Hussain Salleh, Arief R. Harris, J. M. Najeb, and I. KamarulafizamCenter for Biomedical Engineering, Transport Research Alliance

Universiti Teknologi Malaysia, Johor, Malaysia

Abstract— This paper describes the implementation of low cost standalone Digital SignalProcessing (DSP) hardware for ECG monitoring in order to assist physician or cardiologist tocarry out their visual observation in ECG monitoring. The visual observation is in real time,where the acquired ECG from the subject can be seen on the graphic LCD and it is important toensure the signal is stable and consistent before the pre-recording session take place which wouldhelp the cardiologist’s or physician’s in analysis and diagnosis, accurately. The design of the DSPhardware based stand alone system are divided into three hardware units, first the digital signalprocessor with memory unit, second is the signal monitoring displays unit and third is analogfront end unit. Firmware development in this system design is to attach each hardware unit toperform specific functions in carrying out operation on the stand alone system. The advantagesof this system are the ability to run in stand-alone operation and as well as eligibility of updatingon board pre-program algorithm for ECG analysis which existing ECG monitoring equipmentare lacking, which means that this stand alone system could be programmed to suit a particularneed.

1. INTRODUCTION

Non-invasive ECG approach has become the standard practice in clinical environment, after it isfirst made known by Eithoven [1]. Then, the discovery of ECG brings forward the opportunitiesfor further research and development. So, in year 1947, modern ECG recording machine is intro-duce by Norman Jeff holter, later it is called Holter Ambulatory Electrocardiography [2]. Hence,along the path it gave birth to the ECG recording system which is populated by the invention ofmicroprocessor in the later years. Indeed, ECG recording trend is turning a new leaf when mi-croprocessor is embedded with electronic gadget in various hardware platforms, which allows fullassessment to the collected data of electrocardiography [3]. However, in the early microprocessorsystem, the ECG signal is compressed while the system is in recording mode, then it is stored andthe analysis is done on computer [4]. After that, ECG recording also have had evolved throughoutthe decades with the growth of technology in signal processing [5]. Later on, the impact of the sig-nal processing achievements created a paradigm shift and alters the being of hardware computationwhen mathematically algorithms are applied in signal analysis.

Since then, the algorithmic activities show favors in non-computer dependant operation withefficiency in numerical processing. Then, it is followed by the influential of its outcome which isushering the microprocessor step into higher stage by means of doing solo numerical computing.Follow by the pro-active influent, leads the conventional microprocessor to cushion the successorwhich is succeeded with a greater benefit of achievements when the digital signal processor (DSPs)is introduced and in promotion in year 1971. Thus, the DSPs offers built in and integrated withmathematical capabilities and proficiency of numeric functional [6] as compared with others avail-able non-DSP terms in the general microprocessor. Nevertheless, it is also capable of doing “onchip” analysis, which is, including real time signal processing [7]. Therefore, the advantages ofDSPs had open door to enable ECG analysis to be done in real time processing instead of recordingand stored for later analysis.

The introduction of DSP processor had brought forward realization for integrating DSP boardto computer based for real time ECG analysis [8]. One of the common hardware tools involvedECG in computing is the plug n play DAQ card [9]. In this system, add on hardware such as dataacquisition card is required for this purpose. Where the dependency for plug in data acquisitioncard in the computer itself required software development, such as firmware and application soft-ware is necessary to establish communication link with the connected card. Those integrated dataacquisition card with software application can provide a magnificent outcome, but lack of mobil-ity, expensive and computer dependant. But a custom build Printed Circuit Board (PCB) withspecialty electronics design are more prefer to use in stand alone operation for specific task [10].

1902 PIERS Proceedings, Kuala Lumpur, MALAYSIA, March 27–30, 2012

However, few drawbacks of connectivity to computer itself are power dissipation, bulky, complexityand waste of resources to perform only single task purpose [11, 12].

By adapting the idea of custom build and stand alone for specific task operation as suggestingin [10–12], we propose an idea based on single board system design to meet the requirements ofsimplicity, low power consumption, flexibility and low cost. The proposed prototype design is ablefor standalone ECG monitoring system using digital signal processing hardware. Thus, it providesan effective, specific operation as independence signal processing unit to process the acquire ECGin this paper.

Main Board

Remote socket

DSP

USB

ROM

RAM

Microcontroller

GLCD

Codec

SCC

ADC

Analog Front End unit

Signal monitoring

Signal processing

with

Memory unit

a

b c

P

Q

R

S

T

d

Figure 1: Block diagram of prototype DSP hardware stand alone system with the tapped ECG is checkedwith oscilloscope and display on the graphic LCD. The tap ECG also captured by the DSP processor anddisplayed on the window at Code Composer 3x/4x.

2. HARDWARE PLATFORM

Figure 1 shows the DSP hardware based stand alone system design is divided into three functionalunits. The solid lines with arrows symbol represent the flow of data and information. The dashedlines with green, red and blue color represent the functional hardware units. The DSP hardwareconsists of three functional hardware units, they are;

a) Signal processing with memory unit; this unit provides signal altering, changing, re-solving, modification, storage and communication protocol. The mains components are floatingpoint DSP processor (TMS320VC33), Electrical Erasable Programmable Read only Memory —EEPROM (CAT28LV64W) and Static Random Access Memory — SRAM (CY7C1041DV33). TheSRAM memory is expandable in both the data bus and address bus width.

b) Signal monitoring displays unit; this unit provides input signal observation throughthe input peripherals. In this unit, the microcontroller, PIC18F452 is configured to interact withgraphical liquid crystal displays (GLCD). The ADC chip, ADC0820 received input from the tapECG signal through signal conditioning circuit (SCC), then the converted ADC data is sent to theoutput port (DB0 to DB7) of ADC0820.

c) Analog front End unit; this unit provides interaction through peripherals connectivity withthe real world environment. The peripherals such as in Figure 1 shows that the ECG is deliveredto the main board through the custom build AFE device. This AFE input consists of; i) Signalconditioning circuitry (SCC), ii) Codec chip, PCM3003 and iii) Analog to digital converter chipset, ADC0820.

In this DSP hardware design, is utilizing two processors by means of putting a single DSPprocessor (lack of tasks’ oriented but capable of mathematical and algorithm executions) and asingle microcontroller (performs best in tasking, but lack of DSP terms) into one system board.Therefore, this will provide separate traffic management for tasking and signal processing without

Progress In Electromagnetics Research Symposium Proceedings, KL, MALAYSIA, March 27–30, 2012 1903

(a) Test Subject: Resting in static condition

(b) Test Subject: Resting in non-static condition

Figure 2: Real time ECG observation from test subject is captured and displayed on the graphic LCD.

over-loading each processor’s work load. This approach is depicted in Figure 1, the microcontrolleris maintaining to display the tap ECG from ADC chip to the graphic LCD and the DSP processoris focused on reading the tapped ECG signal from the codec chip. However, the Universal SerialBus (USB) in this DSP hardware design is for downloading the ECG analysis algorithm to theROM and it is also use for power up the DSP hardware.

3. RESULTS

The operation results for Analog Front End (AFE) unit interface with volunteer subject as shown inFigure 1(a), Figure 1(b) and Figure 1(c) respectively. Figure 1(a) shows the tapped ECG from testsubject, it is tapped from the Lead II formation on human body skin surface with four disposableelectrodes placed on the right arm (white color), left arm (black color), right leg (green color) andleft leg (red color), all the color code were referred to electrodes placement on human body [13].Figure 1(b) shows the oscilloscope screen display from the output of AFE, this means that it isfunctional. Then, in Figure 1(c) shows that the output from AFE is able to read by microcontroller,PIC18F452 and display the tap ECG to the Graphic LCD.

Emulator XDS510PP (PLUS) from Spectrum Digital Incorporated was attached to on boardJTAG connector at TMS320VC33 by communicate through Code Composer 3x/4x in order tovisualized the tap ECG at the graphical windows as depicted in Figure 1(d). The capacity ofthe graphical windows can only display 2000 data. As a result, 7 cycles of tap ECG data werecaptured in approximately 7 second, all the captured signal were labeled with its P, Q, R, S and Twaves. These collected tapped ECG data are temporary stored in the external RAM as shown inFigure 1(d). Following, in Figure 2 shows the ECG signals monitoring for volunteer subject eitherresting in static condition or non static condition.

4. DISCUSSION AND FURTHER WORK

In common, the graphic LCD has become the stepping stone along the path for portable devicedevelopment and it is important for heart rhythm observation on particular subject with routineheart diagnosis. Therefore, the usage of graphic LCD in Figure 2 has significant impact for real timeECG observation; where this will give better preparation and signal stability during pre-recordingECG session begin on the particular subject. Since, with the proposed real time ECG observation,good ECG recording will lead to accurate ECG analysis and diagnosis significantly.

5. CONCLUSION

We have presented the DSP hardware based stand alone system design by using digital signalprocessor as the core processing unit. The DSP hardware is able to visualize the tap ECG on thegraphical LCD (GLCD) as well as the digital signal processor is able to capture the tap ECG asshown on the graphical window in Code Composer 3x/4x. Nevertheless, this system design haslaid the ground work for stand alone operation in ECG monitoring purposes by using digital signalprocessor for real time signal processing. The hardware system is ready to run any processing

1904 PIERS Proceedings, Kuala Lumpur, MALAYSIA, March 27–30, 2012

algorithm related to ECG. The algorithm can be continuously changed and tested to improvesystem performance and efficiency.

REFERENCES

1. Dijk, J. and B. Van Loon, “Scanning our past from the Netherlands: The electrocardiogramcentennial: Willem Einthoven (1860–1927),” Proceeding of the IEEE, Vol. 94, No. 12, 2182–2185, 2006.

2. Barold, S. and S. J. Norman, ““Jeff” Holter — Father of ambulatory ECG monitoring,” Journalof Interventional Cardiac Electrophysiology, Vol. 14, No. 2, 117–118, Springer, 2005.

3. Kara, S., S. Kemaloglu, and S. Kirbas, “Low-cost compact ECG with graphic LCD and phono-cardiogram system design,” Journal of Medical System, Vol. 30, No. 3, 205–209, Springer, 2006.

4. Watts, M. P. and D. B. Shoat, “Trends in electrocardiograph design,” Journal of the Institutionof Electronic and Radio Engineers, Vol. 57, No. 4, 140–150, IEEE, 1987.

5. Sandham, W., D. Hamilton, P. Laguna, and M. Cohen, “Advances in electrocardiogram signalprocessing and analysis,” Journal on Advances in Signal Processing, Vol. 2007, No. 2007, 1–5,Hindawi Publishing Corporation, 2007.

6. Frantz, G., “The DSP and its impact on technology,” Processor Design: System-on-ChipComputing for ASICs and FPGAs, 101–119, Jari Nurmi, Springer, 2007.

7. Gorup, Z., D. Stajer, and M. Noc, “Signal analisis system for optimal timing of electricaldefibrillation,” 10th Mediterranean Electrotechnical Conference, MELECON 2000, 694–697,2000.

8. Silva, J., J. R. Caldera, and G. Passariello, “Dsp application for QRS detection, alignmentand averaging In ECG signal,” Proceedings of the 1995 First IEEE International CaracasConference On Digital Devices, Circuits and Systems, 149–153, 1995.

9. Low, Y. F., I. Binti Mustaffa, N. Binti Mohd Saad, and A. H. Bin Hamidon, “Developmentof PC-based ECG monitoring system,” 4th Student Conference on Research and Development,66–69, Selangor, June 27–28, 2006.

10. Trusov, A. A., I. Chepurko, A. R. Schofield, and A. M. Shkel. “A standalone programmablesignal processing unit for versatile characterization of MEMS gyroscopes,” IEEE Conferences,2007.

11. Tay, E. J., C. F. Lim, and Hemalatha, “DoIP interfacer system: A low-cost alternative to com-puter for basic network communication in LAN environment,” Journal of Computer Science,Vol. 5, No. 7, 471–475, 2009.

12. Ray, B., “PC Based Instrumentation versus standalone,” IEEE Conferences, 1988.13. Garvey, J. L., “ECG techniques and technologies,” Emergency Medicine Clinics of North Amer-

ica, Vol. 24, No. 1, 209–225, Elsevier, 2005.