Abstract: This paper presents an ECG system using Two Electrodes, where only two electrodes are used for the measurement of the heart signal. This is a portable biomedical instrument which is used as diagnostic tool in every part of the world. As this is Portable, it can be used by a normal person for the regular check up. Here the design has been presented, where the signal from the heart is detected by the electrodes and bootstrapped. This signal is further fed in to ECG amplifier and the output is phase shifted and again fed back to the input of the amplifier to reduce the common mode voltage. The analog signal is filtered and digitized for the interfacing purpose. This design mainly can be used for different types of applications such as ECG Monitors, Holter Monitors and even for telemedicine applications.
Keywords: Electrodes, Instrumentation amplifier, filters,
ECG is a diagnostic tool, used for the measurement of the biopotential signal produced by heart. These biopotential signals are produced by the pumping of blood in circulatory system of the heart .Earlier the ECG instruments were made use of many electrodes (Minimum of Three Electrodes), where two electrodes applied to both the hands and the third electrode is given to the ground. This design doesn't give the portability and feels more uncomfortable to the patients . In this modern world patient needs a portable biomedical instrument where he can diagnose himself . For this, an improved design instrument is required where it is compatible with the computers. The presented design shows that only two electrodes are used and the third electrode which is used as ground electrode is eliminated. This elimination has some advantages like reduction of leakage currents and cost . Now before proposing a model we will study the measurement of ECG.
The flow of blood through every part of the heart produces the contraction of the muscles. This contraction results in production of the electrical impulses, which certainly behave as a voltage source and generates a flow of current
. This current flow gives the activity of the heart, which can be studied by applying the electrodes on different parts of the body. Earlier when the first ECG instrument was introduced twelve electrode were made used, these electrodes measured the electrical activity from different angle. A proper and exact signal can be measured by connecting three electrodes, where the two electrodes will be applied to both the arms and third one is given to ground. The signal or the output waveform can be seen as shown in fig 1.
Now by analyzing this, I propose a model where the ECG signal can be generated by using two electrodes.
Analog Front End
The model is divided in to two parts 1) Analog front end 2) Interfacing. Here we will be concentrating on the analog part where the electrical signal produced from the heart is sensed by the electrodes and fed it to the amplifier for the amplification. The amplifier design has been modeled in such a way that it exempts the need of third electrode. Finally filtering is done for the removal of noise.
Electrodes are applied to sense the electrical impulses which are in terms of milli volts produced from the heart. These act as transducers where the ionic flow from the body is converted to the electronic current with the help of electrolyte . For better contact with the skin, a coupling gel is used and now-a-days dry electrodes are also used for the better performance. The electrode has a sensor which is capable of detecting the signal in the range of 0.5 to 10mv .
Analog front end has got two stages of amplifiers, where both stages have different gain factors. The first stage of amplifier is two input differential amplifier and has a unity gain. The reason for setting the unity gain is because of polarization of electrodes potential .
Bootstrapping is done because to improve the input impedance of the input signals. The op amp's dc bias current corridor is bestowed by these amplifiers . To have the lofty efficient circuit impedance at 60 Hz, a 1M ohm resistors should be driven by feedback capacitor . Followed by this an AC decoupled capacitor is used to prevent the certain noise from the amplifiers.
D. ECG Amplifiers
ECG amplifier is the second stage amplifier of the circuit. This design is responsible for general performance of the signal acquisition system  and should have the capability of amplifying the signal. Instrumentation amplifier is used for the amplification of the signal. It has got two differential input amplifier which is normally called as buffers and one single ended output amplifier.
INA128 is used because of its high CMRR (common mode rejection ration) which is nearly about 120 dB. The other special features of INA128 are, it has got very low DC offset voltage and it can bear up the voltage up to 40V . The gain of this amplifier is set higher than the first stage amplifier and the gain equation is given by
Here the inputs Vin+ and Vin- takes the signal from the first stage of the amplifier and amplifies the signal. This IC is powered up by 5v battery and has got two grounds (one is Ref and another is normal ground). The common mode voltage which is present between the two electrodes will be taken out from the gain resistance inputs.
C. Common mode voltage Reduction
Electromagnetic Interference (EMI) is common voltage that is present in the signal. When the biopotential signals are recorded by the amplifier, the voltages of both amplifiers and patient's are common then it is called as a common mode voltage .
Although INA128 has got the great CMMR but there will be some common voltage left in the signal. To minimize the signal it is fed to the inverting input of the op amp where the signal is inverted. We are using OPA2335 Op amp because it has tolerance capacity of at least 3A of common mode current present in input signal . It has got the other features which are same as the INA128 op amp, like high input impedance and low offset voltage.
E. VOLTAGE TO CURRENT SOURCE
The output of the op amp which has the common mode voltage (fig no 4) is applied to the inputs of the voltage to current source converters. Here, two voltage to current source converter are used which is mainly for two electrodes purpose. It also acts as an amplifier where the current signal is amplified because it should have the current value same as the output of the first stage amplifier. The output of the V to I source is summed up to the differential input of the amplifier and the common voltage which is present in both the electrodes get cancelled. Thus we can attain the differential input impedance like frequency independent common mode and frequency dependent differential input impedance . The potential of polarization effects which are produced in the two electrodes are well adjusted by regulating the input impedance of the input of the amplifier.
Noise is a main factor for tracing an ECG signal. There are several noises that can be interfered in to the signal while recording, such as AC Mains Interference, Motion artifacts and the noise from the surrounding walls. A 50Hz mains supply is a frequency of the supply line which is a noise source for the patient . This supply has got 50Hz of noise content which is added to the ECG signal at the time of recording. The other source of noise is motion artifacts of the electrodes. The movement of the electrodes when applied to the skin gives the 200mv of noise, mainly due to the respiration and perspiration. The other noise which can be interfered is noise from the surrounding walls from the patient room. These noises can be reduced by filtering the signal which filters out all the noise content in the signal. We have used decoupling capacitor at the first stage amplifier which has reduced some amount of noise content in the signal. But the signal is further amplified by the instrumentation amplifier where the noise also gets amplified. To remove the amplified signal we made use of Band pass filters. The Band pass filter is constructed by cascading the low pass filter and the high pass filter. Here the frequency range for measuring the ECG signal used is 0-150Hz . The low pass filter is designed by using RC filters where it passes the low frequency signals and attenuates the frequencies higher than the cutoff frequency. It is designed at the cutoff frequency of 150Hz. By assuming the value of capacitor we can calculate the value of the resistor. The formula used for calculating the values is
fc = 1 / 2pRC
Now the high pass filter is cascaded to the low pass filter where it passes all the high frequency signals and attenuates the frequencies lower the cutoff frequencies. The high pass filter is designed at the cutoff frequency of 0.04Hz where it attenuates signal below this frequency. To calculate the value of the resistor we use the same formula that has been used above.
Isolation is mainly done for the patient safety purpose where it reduces the leakage currents flowing from the instruments to the human body . There are two types of isolation namely electrical isolation and optical isolation. In electrical isolation transformers are used but this is a high freuency device, so biopotential signal needs to modulate using the high frequency carrier. Optical isolation is another isolation where the electrical signal is converetd to optical signal and back to electical signal with the help of LED and detector. In this design we are making use of IC NEC PS2506 as an opto coupler. It has large block voltage mainly in terms of 5000 v r.m.s, good AC input response.
III Analog to Digital conversion
The analog circuitry produces the analog ECG signal, so to have an interface to the PC we need to digitize the signal. The Dspic Microcontroller is used for processing the analog ECG signal which is produced from the analog front end. This microcontroller has got 7 ADC channels of 12 bit, UART pin and is capable of doing the filtration in digital format . As it is having 7 ADC channels but all are multiplexed to one converter, which means that only one channel can be converted at a time. These converted samples are stored in a 16 word buffer between the interrupts. To have a greater efficiency, the interrupt routine will be generated after all the samples are written to the buffer. To convert the signal it must satisfy the Nyquist criteria  which states that, the sampling frequency should be greater than the twice of maximum frequency i.e. fs >2.fmax. As the maximum frequency of ECG signal is 200 Hz, so we need to set the sampling frequency to 400 Hz but the sampling frequency of ADC is set to 6 KHz because to have the filtration of the analog ECG signal. First the signal will be analyzed by accepting the analog ECG signal and then it is processed to the amplifier where we get the correct representation of the ECG signal . The output of the amplifier will be analyzed by the converter and it starts to convert the analog signal in to the digital format. After the conversion of the signal the samples will be stored in the 16 word buffer which is divided in to two channels. The averaging between these two channels will be done by the interrupt service routine and the quantized value will be sent for the filtration of the signal. Although analog filter has filtered out the noise content in the ECG signal but the digital filters are used to remove the noise in digital domain. The digital filters are implemented using MATLAB and tested in the microcontroller. Mainly Digital low pass and notch filter will be used to remove the noise from the mains power supply.
A. Transmission of Signals
The transmission of the signals will be done with help of USB where the serial transmission of signals takes place. The microcontroller used for the ADC has got USART pin at the peripherals through which the data will be transmitted serially . A particular transmit and receive program will be written to send and receive the data via USB IC FT232. This IC has inbuilt USB protocol and there is no need to program it. A specific driver will be installed in the pc and communication will be taken place via USB . To Show the ECG signal on the PC a specific GUI is employed and the visual basic 6 is used for the implementation of the GUI. To have the communication between the USB and to the pc MS Comm ActiveX control is used. In this way data is transmitted serially and displayed on the.
IV. Expected Results
The below figure 1 shows the analog ECG signal where we used two electrodes to generate the ECG signal. A typical amplifier and a bidirectional voltage to current source circuitry were used to remove most of the common mode voltage present in the ECG signal. The frequency ranges of 0-150Hz were used for the measurement of the signal. The output should be amplified up to certain extent and it should not contain any noise content in that. In the fig7 it shows that the signal is digitized and interfaced to the pc. The analog signal sampled out and digitized with the help of the ADC and several digital filters were used to remove the noise content in the signal.
The model has been presented and could be able to trace the trace the signal. It's been shown that the design which has got different values and parameters works fine in order to record the signal. A special care has to be taken for the patient while recording, so optocoupler has been used. The signal is well sampled out in ADC and many of the digital filters is being used in order to remove the noise content in the signal. Present has shown that only two electrodes could be able to use to trace the signal but in the future it can be predicted that we could be able to record the signal with the help of only single electrode.
- Nitish V.Thakor and John G.Webster "Ground free ECG recording with two electrodes", IEEE TRANS.Biomed.Eng.,vol.BME-27,pp-699-704
- D.Dobrev"Two electrode low supply voltage electrocardiogram signal amplifier", Med and Bio Eng, vol 42,pp-272-276
- Patrick O.Bobbie Chaudary, Zeeshan Arif, Hema Chaudari Sagar Pujari "Electrocardiogram (EKG) Data Acquisition and Wireless Transmission" Paper, southern polytechnic state university, USA.
- Lu, M., The Design and Construction of an ECG Telemetry System, M.S. Thesis, University of Queensland, 1994.
- R. P. Betts and B. H. Brown, "Method for recording electrocardiograms with dry electrodes applied to unprepared skin," Med. Biol. Eng., vol. 13, pp. 313-315,1976.
- Mihai Rqmanca Gheorghe Pana Iuliu Szemely Florin Sandu" A low nosie instrumentation amplifier for ECG recording", IEEE TRANS, vol 3 pp-707-710.
- Spinelli, E.M. Martinez, N.H. Mayosky," A transconductance driven-right-leg circuit ",IEEE TRANS, Biomed Eng, vol 46,pp 1466-1470
- R.Khandpur "Biomedical Instrumentation", Publisher Tata Magraw Hill.
- J. H. Nagle, "Biopotential amplifiers," in The Biomedical Engineering Handbook, Bronzino J. D.,Ed., Boca Raton, FL: CRC Press, pp. 1185-1195, 1995.
- Nitish V. Thakor. "Bipotentials and Electrophysiology Measurement.": CRC Press, Available: http://www.scribd.com/doc/6938892/Biopotential-Measurement
- Open ECG project "3 Lead wireless ECG", Available: www.open-ecg-project.org