Development of the injection current source for the information-measuring system of electrical impedance tomography of biological object

04023


Introduction
One of the main components of an electrical impedance tomography (EIT) device, the characteristics of which affect the stability and metrological characteristics of the entire device, is the injected current source.
The developed CS injected current source for the EIT device must meet the following requirements: -control of the shape, frequency fI and amplitude Im of the current by control voltage Ucont ; -amplitude Im output current I through load Rload : 1-5 mA [1]; -error δI amplitude Im output current I -no more than 5%; -maximum frequency fImax -100 kHz; -grounded load with resistance Rload 50 Ohm -2 kOhm [1].
These requirements are met by an CS circuit with automatic measurement and regulation of current in the load [2,3].

Materials and methods
The functional diagram of the developed source for the EIT device is presented in Figure (1).The output current I is directly proportional to the voltage drop Us across the resistor R s and inversely proportional to the resistance of the resistor Rs.To measure the voltage drop Us, a differential amplifier is used on the operational amplifier DA2.If the gain of the differential amplifier [4]

Simulation of a current source circuit
Schematic electrical diagram of the developed CS for the EIT device in the Micro-Cap 10.0 circuit modeling package [5] is shown in Figure (3).The GENERIC model was selected as an op-amp with parameters corresponding to the parameters of the Texas Instruments TL 07 CP op amp [6].  1. Im of the current I through the load Rload was calculated.Next, the calculation of the reduced δI error in establishing the amplitude Im of the current I was carried out.The calculation is made according to the formula: where  As can be seen from Figure 4, as the value of the load resistance Rload increases, the error δI of the amplitude Im increases output current I.This is caused by the influence of the load on the current-generating circuit.The minimum error δI is observed at Rload ≈ Rs.With increasing frequency fcont of the control signal Ucont, the error δI of the amplitude Im output current I also increases, but the influence of frequency fcont is significantly less than the influence of the resistance value Rload .The increase error δI with increasing frequency fcont is caused by the imperfection of operational amplifiers.According to the simulation results, the error δI of the amplitude Im output current I does not exceed 1% at resistance values Rload = 50.. 100 Ohms and frequency fcont = 1..100 kHz.With load resistance Rload = 0.5..1 kOhm and frequency fcont = 1..100 kHz error δI amplitude Im output current I does not exceed 5%.

Making a current source
The electrical circuit diagram of the developed CS injected current source for the EIT device is shown in Figure (2).
When power is connected, all EIT device units are supplied with supply voltage.The shape and frequency fI of the current I at the OUT pin of the current source correspond to the shape and frequency fcont of the control signal Ucont at the IN pin of the current source.The amplitude Im of the current I at the CS output is proportional to the amplitude Um control voltage Ucont at the IN CS pin .The current source starts working immediately after power appears.IN CS pin is connected to the common point through resistor R1 to avoid the appearance of current I at the OUT pin of CS due to the effect of induced voltage on the IN pin of CS.For selected operational amplifiers DA1, DA2, the maximum output voltage is at least ±10 V A drawing of a printed circuit board has been developed and an assembly drawing of the CS injected current source for the EIT device is shown in Figure (5).Development carried out in DipTrace software [7].

Results and discussion
In experimental studies of CS for EIT device, it is necessary to check the following characteristics: -error δf of output current I frequency fI; -error δI output current I amplitude Im.
To experimentally estimate the error δf of the frequency fI of the output current I, we will estimate the frequency fload voltage Uload at the reference load Rload.Load resistance Rload = 220 Ohm.To measure frequency, a frequency meter FM-63 is used.The main characteristics of the device are summarized in table (2).To generate the control signal Ucont, a special form signal generator AKIP-3409/5 is used [8].The main characteristics of the generator are summarized in table (3).
The error δf of the frequency fI of the output current I will be calculated using the formula:  The results of assessing the error δf of the frequency fI of the output current I are presented in Figure 7.The figure shows the arithmetic mean value of the error δf for each value of the frequency fI of the output current I indicating the standard deviation.As can be seen from Figure 7, the maximum relative error δf of the frequency fI of the output current I does not exceed 0.07%.Starting from frequency fI = 8 kHz error δf does not exceed 0.001%.
To experimentally estimate the error δI of the amplitude Im output current I we will estimate the amplitude Im current I through the reference load R load .The resistance store P4831 [9] is used as the load Rload.Load resistance Rload = 50;100;500;1000;1500;2000 Ohm.The main technical characteristics of the resistance store are presented in table (4).To measure the amplitude Im of the current Iload , a universal voltmeter AKIP-2101 is used [10].The main characteristics of current measurement with the AKIP-2101 voltmeter are summarized in table (5).
The AKIP-2101 voltmeter allows you to measure the effective IRMS value [10] of the current, taking into account the signal shape and distortion (True RMS).The amplitude Im value is calculated using the formula: where Im is the amplitude of alternating current; I RMS -effective value of alternating current.
Calculation of the error δI of the amplitude Im output current I will be produced according to the formula: where Im _meas is the measured value δI of the amplitude Im output current I; Im -specified value δI of the amplitude Im output current I, IМ -maximum specified value δI of amplitude Im output current I.
The experiment is carried out for frequencies fcont = 1; 10; 20; 30; 40; 50; 60;70; 80; 90;100 kHz and amplitudes Im =1; 2; 3; 4; 5 mA.The number of experimental experiments was 11 repeated experiments.The diagram and appearance of the experimental stand are presented in Figure 8.As can be seen from Figure 9, as the value of the load resistance Rload increases, the error δI of the amplitude Im increases output current I.This is caused by the influence of the load on the current-generating circuit.With increasing frequency fcont of the control signal Ucont, the error δI of the amplitude Im output current I also increases, but the influence of frequency fcont is significantly less than the influence of the resistance value Rload.The influence of frequency in the experiment is more obvious than in the simulation.According to the experimental results, the error δI of the amplitude Im output current I does not exceed 5% at resistance values Rload = 50..500 Ohm and frequency fcont = 1..50 kHz.With load resistance Rload = 1 kOhm and frequency fcont = 1..30 kHz, amplitude error δI of Im output current I does not exceed 5%.With load resistance Rload = 1.5 kOhm and frequency fcont = 1..10 kHz error δI amplitude Im output current I does not exceed 5%.
The human forearm was also used as a load.To do this, using a special medical belt, two rheographic electrodes were attached approximately in the middle of the forearm.The electrodes were located diametrically opposite.The appearance of the CS connection to the BO is shown in Figure 10.To improve contact between the electrodes and the skin, pieces of fabric moistened with water were placed.The experiment is carried out for frequencies fcont = 1; 10; 20; 30; 40; 50; 60; 70; 80; 90;100 kHz and amplitudes Im =1; 2; 3; 4; 5 mA.

Conclusion
As a result of the work, an CS injected current source for the EIT device was developed, assembled and investigated.This current source is controlled by voltage and allows you to pass through a grounded load Rload up to 2 kOhm a current with an amplitude Im of up to 5 mA.Study of the error δI of the amplitude Im output current I and error δf frequency f I output current I was produced on a mathematical model and during a full-scale experiment.Active resistance and a biological object were used as the load Rload.As a result of the simulation, it was established that the error δI of the amplitude Im output current I does not exceed 1% at resistance values Rload = 50..100 Ohm and frequency fcont = 1..100 kHz.With load resistance Rload = 0.5..1 kOhm and frequency fcont = 1..100 kHz error δI amplitude Im output current I does not exceed 5%.As a result of a full-scale experiment, it was established that when using active resistance as a load Rload, the error δI of the amplitude Im output current I did not exceed 5% at resistance values Rload = 50..500 Ohm and frequency fcont = 1..50 kHz.With load resistance Rload = 1 kOhm and frequency fcont = 1..30 kHz error δI amplitude Im output current I does not exceed 5%.With load resistance Rload = 1.5 kOhm and frequency fcont = 1..10 kHz error δI amplitude Im output current I does not exceed 5%.However, this error can be compensated by changing the amplitude U m control voltage Ucont.To do this, it is necessary to introduce an exemplary resistance into the load circuit and measure the voltage drop across it.Based on the measured data, it is possible to calculate the amplitude value Im output current I and compensate for the error δI.As a result of a full-scale experiment when using a biological object as a load Rload, the amplitude Im error δI of the output current I does not exceed 5% at frequency fcont = 10.. 30 kHz.

Fig. 2 .
Fig.2.Schematic diagram of CS EIT device When power is connected, all EIT device units are supplied with supply voltage.The shape and frequency fI of the current I at the OUT pin of the current source correspond to the shape and frequency fcont of the control signal Ucont at the IN pin of the current source.The amplitude Im of the current I at the CS output is proportional to the amplitude Um control voltage Ucont at the IN CS pin.The current source starts working immediately after power appears.Current source IN pin is connected to the common point through resistor R1 to avoid

Fig. 3 .
Fig.3.Model of CS EIT device in the Micro -Cap environment Based on the simulation results, an estimate was made of the error δ I of the amplitude Im of the current I through the load Rload under various CS operating modes.The model parameters simulating various CS operating modes are summarized in Table1.
_ mod is the calculated value δI of the amplitude Im output current I; Im -specified value δI of the amplitude Im output current I, IМ -maximum specified value δI of amplitude Im output current I. Estimation of the spread of error values δI of amplitude Im output current I depending on the load resistance Rload is presented in the form of a Box-Whiskers diagram in Figure (4).The diagram was built using the Statistica package.

Fig. 4 .
Fig.4.Estimation of error values δI for different values of Rload and fcont Estimation of the spread of error values δ I of amplitude Im output current I depending on the frequency fcont of the control signal Ucont is presented in the form of a Box-Whiskers diagram in Figure (4) too.As can be seen from Figure4, as the value of the load resistance Rload increases, the error δI of the amplitude Im increases output current I.This is caused by the influence of the load on the current-generating circuit.The minimum error δI is observed at Rload ≈ Rs.With increasing frequency fcont of the control signal Ucont, the error δI of the amplitude Im output current I also increases, but the influence of frequency fcont is significantly less than the influence of the resistance value Rload .The increase error δI with increasing frequency fcont is caused by the imperfection of operational amplifiers.According to the simulation results, the error δI of the amplitude Im output current I does not exceed 1% at resistance values Rload = 50.. 100 Ohms and frequency fcont = 1..100 kHz.With load resistance Rload = 0.5..1 kOhm and frequency fcont = 1..100 kHz error δI amplitude Im output current I does not exceed 5%.

Fig. 5 .
Fig.5.Appearance of assembled CS for EIT device where fmeas is the measured value of the frequency f I of the output current I; fcont -set value of frequency fI of output current I, fМcont -maximum set value of frequency fI of output current I.The experiment is carried out for frequencies fcont = 1;10;20;30;40;50;60;70;80;90;100 kHz.The number of experimental experiments was 11 repeated experiments.The diagram and appearance of the experimental stand are presented in Figure 6.

Fig. 7 .
Fig.7.Estimation of error δf frequency fI of output current I

Fig. 8 .
Fig.8.Scheme (a) and appearance (b) of the experimental setup After calculating the error values δI amplitude Im output current I for each series of repeated experiments, the arithmetic mean value of the error δI was calculated and processing was carried out similar to the processing of the simulation results.Estimation of the spread of error values δI of amplitude Im output current I depending on the load resistance Rload is presented in the form of a Box-Whiskers diagram in Figure (9).The diagram was built using the Statistica package.

Fig. 9 .
Fig.9.Estimation of error values δI for different values of Rload and fcont Estimation of the spread of error values δI of amplitude Im output current I depending on the frequency fcont of the control signal Ucont is also presented in the form of a Box-Whiskers diagram.As can be seen from Figure9, as the value of the load resistance Rload increases, the error δI of the amplitude Im increases output current I.This is caused by the influence of the load on the current-generating circuit.With increasing frequency fcont of the control signal Ucont, the error δI of the amplitude Im output current I also increases, but the influence of frequency fcont is significantly less than the influence of the resistance value Rload.The influence of frequency in the experiment is more obvious than in the simulation.According to the experimental results, the error δI of the amplitude Im output current I does not exceed 5% at resistance values Rload = 50..500 Ohm and frequency fcont = 1..50 kHz.With load resistance Rload = 1 kOhm and frequency fcont = 1..30 kHz, amplitude error δI of Im output current I does not exceed 5%.With load resistance Rload = 1.5 kOhm and frequency fcont = 1..10 kHz error δI amplitude Im output current I does not exceed 5%.The human forearm was also used as a load.To do this, using a special medical belt, two rheographic electrodes were attached approximately in the middle of the forearm.The electrodes were located diametrically opposite.The appearance of the CS connection to the

Fig. 10 .
Fig. 10.Appearance of CS connection to BO Figure (11) shows a graph of the error δI versus frequency fcont control signal Ucont at different amplitudes Im of current I when using a biological object as a load.

Fig. 11 .
Fig. 11.Results of calculating the error δIAs can be seen from Figure(11), at low frequencies the resistance exceeds 2 kOhm, which does not allow the CS to deliver the required current.The highest accuracy was achieved with amplitude Im = 5mA and frequency 10 kHz.Error δI of amplitude Im output current I does not exceed 5% frequency fcont = 10..30 kHz.