Technological advancement in bio-signal processing has enabled computer-based analysis of ECG which is an accessory to the doctor, and helps them in characteristic extraction and measuring of ECG signal for diagnostic intents. However, acquisition and dependable parametric quantity extraction of ECG signals utilizing a computing machine based acquisition system is highly sensitive to power line intervention even though its magnitude is unperceivable. Therefore, normally 50 Hz active parallel notch filter is used to cut down the consequence of power line intervention. An effort has been made in this paper to plan different 50 Hz active parallel notch-filter topologies in P-Spice that produce a notch as opposed to band rejection utilizing minimal figure of high preciseness constituents. Simulation consequences of these filters are evaluated for their amplitude response, stage response and group hold to frequency expanses. Based on the simulation consequences, add-on notch filter hardware are developed and used in concurrence with existent clip computing machine based ECG amplifier system. The front terminal hardware of amplifier system includes biosensors, cascaded amplifiers, right leg thrust circuits, active filters and 50 Hz notch filters. Analog end product of the front terminal system is so interfaced to the computing machine for sing. The intent of this work is to develop computing machine based compact and enhanced signal acquisition system to give ECG signal of greatest clinical usage.
Keywords: Biosignal processing, ECG, Notch Filter, Real Time, P-Spice, MATLAB
A patient in propinquity to power coachs gets capacitively coupled through the isolated capacity between them, to the extent that if one square meter of the human organic structure is coupled to the power lines one meter off, so the isolated electrical capacity is of the order 8.85 pF. Therefore, 50 Hz power line electromotive force coupled with isolated electrical capacity, exceeds the bio-potential degree, and interferes with the measuring of ECG signal. Power line intervention may be reduced by utilizing differential amplifier of high common manner rejection ratio ( CMRR ) in planing the cascade informations acquisition unit. Modern biomedical amplifiers have really high CMRR but are still contaminated by power-line intervention. This is due to differences in the electrode electric resistances and isolated currents through the patient and the overseas telegrams. Hardware solutions have been developed to increase the existent CMRR by equalisation of the overseas telegram shield and the right leg thrust circuits [ 1 ] . This reduces the influence of stray currents through the organic structure, but the efficiency obtained is non important.
Enormous attempts have been made to turn to this issue in the yesteryear. Li BF et Als have attempted a self-adapting correlativity method for fast riddance of 50-Hz noise from the ECG signal in PC-based digital ECG recording system [ 2 ] . Dotsinsky I et al assessed the efficiency of notch filters and a minus process for power-line intervention cancellation in ECG signals. However, the hardware used was incapable for power-line frequence cancellation, so a package measuring of the power-line intervention period was developed [ 3 ] . As a package attack, the intervention mention signal and its quadrature, are linearly combined to be subtracted from any one of the channels to cut down line intervention [ 4 ] . Tabakov S et Al described fast calculation methods for real-time digital filtration and QRS sensing in computerized ECG systems for long-run monitoring. Trials made on standardised ECG database were capable in taking baseline rovings and 50 Hz interventions. The public presentation of the QRS sensor had average sensitiveness of 99.65 % and positive prognostic value of 99.57 % [ 5 ] . Further to this, M.S. Chavan et Al used Chebyshev type I and Chebyshev type II digital filters on the ECG Signal utilizing MATLAB FDA tool. ECG information was acquired from the ego designed instrumentality amplifier but, the interfacing of ECG amplifier to the computing machine was done utilizing advantech PCL 711B attention deficit disorder on card [ 6 ] . They besides presented the application of Butterworth and Elliptic notch and high base on balls digital IIR filter on the natural ECG signal in existent clip [ 7 ] . Kaur M et Al combined the traveling norms technique and IIR notch filter to cut down the power line intervention holding fewer coefficients and therefore lesser calculation clip, therefore applicable to existent clip processing. However, the consequences have been concluded utilizing Matlab and MIT-BIH database [ 8 ] . Mbachu C.B worked on FIR digital filters with Kaiser Window to take the power line interventions utilizing FDA tool [ 9 ] . But, additive filters designed to cut down power line interventions are non really effectual and besides have long impulse response which causes big holds in ECG monitoring and analysis. However, the non-linear filters cater to these restrictions and efficaciously cut down these interventions as stated by Leski J et Al [ 10 ] . Their analysis was nevertheless done on man-made signal. Literature besides reports the survey of the effects of AC intervention and its filtering on the truth of bosom rate sensing for bosom rate variableness analysis [ 11 ] .
ECG information is analyzed chiefly based on the form of the PQRST moving ridge, so it is of import to see how the filter changes its form. Analog filters like Wien and duplicate Tee or digital filters like Butterworth, Chebychev and Elliptic can be used but they non merely rarefy a coveted frequence, but affect next frequences as good, therefore altering the basic ECG moving ridge form [ 12 ] . Adaptive filtering introduces unacceptable transeunt response clip, and digital notch filters are computing machine based solutions to this job. Therefore to cut down the 50 Hz noise in a Personal computer based system, a hardware Notch filter can be a solution which wholly attenuates a peculiar frequence and affects none other [ 13 ] .
Therefore, assorted 50 Hz Notch filter designs are tested in PSpice before implementing the filter hardware to be used with the amplifier unit. The front terminal hardware therefore includes biosensors and self developed tested cascaded amplifiers, right leg thrust circuits, active filters and 50 Hz notch filters. Analog end product of the front terminal system is so interfaced to the sound port of the computing machine and can be viewed on a package CRO [ 14 ] . The intent of this work is to understand how modern ECG is derived and displayed and to determine criterions that will better the preciseness and public-service corporation of the ECG in pattern. Emphasis is placed on compact and enhanced signal acquisition system and computer-based signal processing, which provide automated measurings that lead to computer-generated diagnostic illations.
Analysis Of ECG Signal
This subdivision examines the relation of the resting ECG to its engineering and shall assist clinicians to understand the losing nexus between engineering and its effects for clinical ECG reading. ECG presents are recorded by digital, machine-controlled machines equipped with package that measures ECG intervals and amplitudes, provides a virtually immediate analysis, and besides compares the tracing to those recorded earlier by the same system. Automated ECG Signal analysis and processing include signal acquisition, filtering, wave form acknowledgment, characteristic extraction, and diagnostic categorization. This requires equal apprehension of ECG signal frequence scope, its filtering and proper lead arrangement.
Band base on balls filtering in the scope of 1 Hz to 30 Hz removes artifacts from the ECG signal, but distorts its high- and low-frequency constituents and is hence non suited for entering intents. Inadequate high-frequency response may do wrong appraisal of signal QRS extremum amplitude and improper low-frequency response can ensue in deformations of repolarization. Therefore, cut off frequence scope for filtrating ECG acquired on a computing machine must be chosen so that it suppresses low-frequency noise ensuing from baseline wander, motion, and respiration and higher-frequency noise that consequences from musculus artifact, power-line or radiated electromagnetic intervention [ 15 ] . Diagnostic reading of ECG utilizing computer-based system ab initio require signal readying like filtering, and template formation followed by sensing and characteristic extraction. It hence becomes indispensable to understand assorted ECG signal frequence ranges and its consequence for proper filtering. Table I below gives the frequence inside informations of the ECG signal.
Primary moving ridge in ECG signal
1 – 2 Hz
Diagnostic information in QRS ( Adults )
& gt ; 100 Hz
Diagnostic information in QRS ( babies )
250 – 500 Hz
Low-frequency cut-off for diagnostic ECG utilizing regular filters
Does non extinguish the baseline impetus
Low-frequency cut-off for diagnostic ECG utilizing additive digital filters with zero stage deformation
Reduces ST-segment deformations
High-frequency digital filter cut-off in grownups
100 – 150 Hz
Reduces error to about 1 % .
High-frequency digital filter cut-off in babies
Maintain diagnostic truth of ECGTable 1. The ECG Signal frequence
Design & A ; Development of ECG Acquisition System Using Hardware Notch Filters
The design and execution of hardware parallel circuits is nevertheless, complex and computationally intensive. So, computing machine package like P-Spice proves out to be really helpful in virtually planing these filters [ 16 ] . The subdivision below gives the design considerations for 50 Hz parallel notch filters.
3.1. Design of 50 Hertz Notch Filters
Active notch filters have been used in the yesteryear for extinguishing constituents of 50-Hz busyness due to their tremendous advantages over inactive filters and because they provide first-class low frequence filtering, which is the premier demand in ECG signal analysis. Active filters eliminate bulky inductances so require less infinite, are inexpensive, and supply good isolation between beginning and burden with suited addition. In planing these filters care needs to be taken in make up one’s minding the Centre frequence tuning, stableness, and repeatability as selectivity and centre frequence depends on the addition. A figure of notch-filter topologies are hence explored with design ends that:
generates a notch in-place of rejecting set of frequences
provides a good notch deepness
green goodss ideal frequence response, stage response and group hold
utilizations minimal operational amplifiers and high preciseness constituents
allows easy tuning for Centre frequence & A ; Q factor
Using P-Spice, notch filters are made by standard resistances, capacitances and UA741 operational amplifiers, as shown in Figure 3.1 ( a ) , 3.2 ( a ) , 3.3 ( a ) , 3.4 ( a ) , 3.5 ( a ) , 3.6 ( a ) & A ; 3.7 ( a ) . The constituent value inside informations for assorted circuits and computation for centre frequence is detailed in Table 2. Analyzing the filters in P-Spice with a frequence expanse resulted in frequence response secret plans of the filters and is given in Figure 3.1 ( B ) , 3.2 ( B ) , 3.3 ( B ) , 3.4 ( B ) , 3.5 ( B ) , 3.6 ( B ) & A ; 3.7 ( B ) followed by their stage response and clip hold secret plans in subdivisions ( degree Celsius ) and ( vitamin D ) severally. A duplicate T-network is most normally used notch filter, capable of bring forthing boundlessly deep notch. Twin T-network is a inactive web dwelling of 2 T-shaped webs. One T-network is made up of 2-resistor & A ; a capacitance while the other is made of two capacitances & A ; a resistance. The frequence at which maximal fading occurs is called notch out frequence ( denoted as f0 ) and is given in Eq ( 1 ) as:
— — – ( 1 )
The major disadvantage of duplicate T-network is that it has a comparatively low ‘Q ‘ and affects the selectivity of filter. This can be improved by increasing the Q value by utilizing a electromotive force follower whose end product is fed back to the junction of R/2 and 2C. The C value of notch is chosen less than 1 microfarad & A ; so R is calculated from Eq. ( 1 ) for the coveted notch out frequence, which is 50 Hz in our instance ( India ) . In topographic point of electromotive force followings we can besides utilize an emitter follower but advantage of electromotive force follower in that it has much higher input opposition and the end product amplitude is precisely equal to the input in magnitude and in stage. The combination of duplicate ‘T ‘ agreement with an operational amplifier in electromotive force follower manner improves the behaviour farther by bettering the Q factor from 0.3 to little greater than 50. The deepness and frequence of notch remains unchanged by adding a electromotive force follower in the filter circuit. The electromotive force follower acts as a buffer that provides a low end product opposition. The usage of operational amplifier provides high input electric resistance and so it becomes possible to utilize big opposition values and low electrical capacity values in the ‘T ‘ web for low frequence applications.
C ( in nF )
Table 2. Component value inside informations for assorted circuits and computation for centre frequence
Figure 3.1 ( a ) is a Twin-T 50 Hz notch filter constellation where resistance R5 is half in value of resistances R1 and R2. Resistance value of R3 is ideally designed utilizing combinations of R1 and R2. Similarly, capacitance C4 and C5 are half the value of C3. Capacitance value of C3 is obtained by parallel combination of C4 and C5, which are same in value. The circuit in figure 3.2 ( a ) uses two operational amplifiers, but the figure of high preciseness constituents required is more. Figure 3.3 ( a ) is a really simple notch filter design made utilizing a individual amplifier and requires the minimal figure of R and C constituents. In figure 3.4 ( a ) , the junction at R2 and C1 is bootstrapped to the end product of the electromotive force follower. It is a individual op-amp filter circuit that raises the Q in proportion to the signal fed back to R2 and C1. Figure 3.5 ( a ) is a modified version of the notch filter in figure 3.4 ( a ) , where two electromotive force followings are used. A 2nd follower is used to further stabilise the Q factor and allows rejection over a broad input frequence ranges. A fraction of the end product is fed back to the junction at R3 and C1 by the 2nd electromotive force follower. The sum of signal fed back affects the notch Q. A potentiometer can be used in topographic point of the resistance R4 in figure 3.5 ( a ) to hold variable Q.
Figure 3.6 ( a ) is a individual op-amp notch filter circuit that gives a really good notch deepness as compared to other individual op-amp notch circuits. The circuit nevertheless is non really flexible, as accommodations of centre frequence becomes hard. Paring the centre frequence requires accommodations of the three resistances R1, R4 and R5. The circuit requires six high preciseness constituents for tuning, of which two are the ratios of others. It is required that R2 and R3 should be really little as compared to R4 and R5. This raises the spread of resistance values which affects the deepness of notch and the centre frequence. Figure 3.7 ( a ) notch filter circuit is similar to the circuit of Figure 3.4 ( a ) with different values of R and C.
Figures 3.1 to 3.7 ( subdivisions a, B, degree Celsius, vitamin D )
The undermentioned observations can be made from the response curves of these notch filter circuits:
In the pass-band, the magnitude response curve is about level
Phase response alterations in the halt set. Largest alteration is observed at the centre frequence
The group hold is high at halfway frequence of 50 Hz
Notch deepness enormously depends on constituent matching
The deepness of the notch obtainable in simulation consequences is theoretical and is non the deepness that can be achieved with real-world constituents due to their tolerance values. The nonsubjective to plan assorted 50 Hz notch filters for medical monitoring devices is the rejection of a specific interfering frequence. As can be seen in Table 2, the constituent values chosen are such, which produce a notch at the halfway frequence of 50.02 for most of the circuits analyzed.
Comparison is done for assorted theoretical consequences obtained in figures 3.1 to 3.7 sub-sections B, degree Celsius and d. It clearly shows that crisp slumps are non possible utilizing designs in Circuit 2, 3 and 6 of figure 3.2, 3.3 & A ; 3.6 severally. However, the notch deepness accomplishable is singular in these circuits. Circuit designs 4 and 5 of figure 3.4 and 3.5 severally, give crisp cut off notch filters, but the notch deepness is non really good. Circuit of figure 3.7 does non give a good notch frequence. The analysis clearly shows that the consequence of Circuit 1 of figure 3.1, if implemented can give a good notch deepness with crisp cut-off.
Practically, strong fading and reasonably crisp slumps are possible utilizing notch filters. Therefore, this survey supports the fact that parallel notch filters, if constructed based on simulation consequences, can be used as add-on circuit for remotion of 50 Hz busyness in ECG amplifier units. Circuit 1 when analyzed utilizing P-spice gave the best notch deepness and crisp cut-off. Thus this circuit was implemented in the research lab utilizing two 9 V batteries as shown in the figure 3.8. The consequences obtained in figure 3.8 utilizing precise values of Resistors & A ; Capacitors about matched the theoretical consequences and so was used in concurrence with the amplifier system to obtain better bio-signal by cut downing the consequence of power line intervention.
Material And Method
Functional block diagram of the developed enhanced system for ECG acquisition is shown in Figure 4.1. Disposable Ag-AgCl electrodes are positioned on both carpuss and the right leg of the topic after using electrolytic gel to pick the ECG signal.
The ECG signal acquired by these non-invasive electrodes is 1mV extremum to top out so an elaboration of about 500 is done in cascade to do ECG signal useable for sensing of bosom inside informations. Front terminal of amplifier block A1 and A2 in the block diagram of Fig. 4.1 along with blocks A3, A4, A5 and A6 that provides overall addition of 500, improved CMRR and do non acquire saturated [ 14 ] . This is followed by a 50 Hz Notch filter developed utilizing standard resistances, capacitances and UA741 Operational Amplifiers. Further, sound card of the computing machine is used as the interface unit, to expose the signal in MATLAB [ 14 ] .
Result And Discussion
Multiple phase elaboration, proper lead arrangement, right leg thrust circuit, parallel filter, hardware 50 Hz notch filter and MATLAB based practical CRO with digital filter plan helped in obtaining human ECG signal. Figure 5.1 shows the snap shooting of remainder ECG signal without utilizing the Notch Filter. X-axis in the consequence represents the figure of ECG samples acquired w.r.t clip. The trying rate was set to 8000 samples per second in this work. Therefore, the consequences obtained demoing 80,000 samples really represents 10 seconds of informations. Y-axis gives the amplitude of the ECG signal. Interference present in Figure 5.1 has been reduced by utilizing 50 Hz notch filter, which is clear from Figure 5.2.