Electrical impedance tomography system: an open access circuit design
© Soleimani; licensee BioMed Central Ltd. 2006
Received: 08 February 2006
Accepted: 03 May 2006
Published: 03 May 2006
This paper reports a simple 2-D system for electrical impedance tomography EIT, which works efficiently and is low cost. The system has been developed in the Sharif University of Technology Tehran-Iran (for the author's MSc Project).
The EIT system consists of a PC in which an I/O card is installed with an external current generator, a multiplexer, a power supply and a phantom with an array of electrodes. The measurement system provides 12-bit accuracy and hence, suitable data acquisition software has been prepared accordingly. The synchronous phase detection method has been implemented for voltage measurement. Different methods of image reconstruction have been used with this instrument to generate electrical conductivity images.
The results of simulation and real measurement of the system are presented. The reconstruction programs were written in MATLAB and the data acquisition software in C++. The system has been tested with both static and dynamic mode in a 2-D domain. Better results have been produced in the dynamic mode of operation, due to the cancellation of errors.
In the spirit of open access publication the design details of this simple EIT system are made available here.
The imaging of electrical properties of different materials has been the main topic of many investigations for a number year [1–3]. In EIT, the contrasts in electrical properties, i.e. the conductivity distribution inside an object, is used to generate a tomographic image [2, 4]. EIT has potential applications in both medical and industrial fields [5–7]. The advantage of such a technique over more traditional imaging modalities (PET, CT, MRI and ect.) is such that, it provides a non-invasive (or "non-destructive") method and requires no ionizing radiation. Furthermore, EIT is a relatively low cost and simple functional technique. The most significant drawback of EIT is its poor image resolution, which is often restricted by the number of electrodes used for data acquisition. Data acquisition is typically made by applying an electrical current to the object using a set of electrodes, and measuring the developed voltage between other electrodes [2, 8, 9].
The major mathematical modeling of EIT involves calculation of the forward and inverse problems. In the forward problem the governing equations in the EIT field which are derivable from Maxwell's Equations (electrostatic approximation for low frequency)  are
where the Ũ(P) is the voltage and (P) is the specific admittance of B; in which (P) = σ(P) + j ωε(P). Equations (1) to (3) are the basic equations used in developing an algorithm to work in an EIT field.
In order to map the resistivity inside a body in a more efficient way, an EIT system SUT-1 has been fabricated .
For the the I/O module, an ADVANTECH PCL-812PG I/O card is used . It consists of a 16 bit programmable I/O card with a 12-bit successive approximation analogue to digital converter, (30 kHz sampling rate), programmable Time/Counter/Gain and two 12 bit monolithic multiplying digital to analogue converter output channels. Due to the application of an unsophisticated analogue to digital conversion algorithm, it is not a fast sampling card.
In this module a fixed frequency current source was designed. A detailed diagram of such a current driver is shown in figure 4. For an EIT current generator, the amplitude stability and high output resistance are the most important aspect of the design . Different circuits were built and tested, and finally reached a digital generation method by means of an EPROM (27C258). Furthermore, the EPROM was programmed to produce 256 steps of a 23 kHz sinusoidal waveform. An 8-bit counter was used for reading the EPROM data, and then data were applied to a digital to analogue converter (DAC-0808). The system internal clock ran at 6 MHz. One of the most important advantages of this circuit is related to the synchronous pulses for demodulation, which can be obtained by the address line decoding. Zero crossing point and amplitude peak point can also be determined. The total harmonic distortion (THD) of this current generator is determined to be about 1.3%. The output of this digital oscillator is fed into the current source through a normal gained buffer stage (LF-357). It must be noted that the voltage control current source (VCCS) is a buffered current mirror circuit. We use Analog Devices AD644 are used as the main part and some LF-411 and LF-412 for buffering. The output current is not more than 5 mA.
In order to perform data acquisition in 16-electrodes and 32-electrodes mode, a multiplexer circuit is necessary for switching the current injector and voltmeter among the different data channels. Our multiplexer circuit (MUX) consists of four 32 × 1 analogue multiplexers. Each multiplexer is a combination of two 16 × 1 IC-4067 multiplexers. The most significant type of errors arising out of the MUX board, labelled as ron, are related to the semiconductor switches, and also cross-talk between different channels. It has to be noted that the ron does not have a constant value, but different values for different channels. It is a function of different parameters such as temperature, current, etc in each channel. It is desirable to have the value of ron as low as possible.
Electrodes and phantom
Different cylindrical phantoms are used in this device. In order to simulate behaviour of the human tissue, saline solutions with different concentrations are used. Normal ECG electrodes i.e. Ag-AgCl type, served as an electrical contacting media. Cu electrodes could be another choice for a better and more realistic simulation of electrode-skin contact impedance [17, 18].
During APT mode of operation, image reconstruction is performed with 16 electrodes using a Back Projection algorithm and iso-potential lines. Basically this was the "Sheffield Algorithm" with some changes and modifications corresponding to the SUT-1 specification.
The system performance was tested with different approaches i.e., simulations and real measurement. Some examples of these results are discussed below.
Results from the simulations
Results from the real measurements
Figure 10 shows the design and experimental results for a phantom with two objects. Measured data were transferred to the computer, the reconstruction algorithm applied and an image was obtained using the back projection method. As seen in Figure 11, a star artifact resulted from the back projection a well-known artifact for this method without using filters. Basically, it is due to limitation in the number of projections. If the projections number (ray-sum) is increased, the size of this artifact will decrease.
SUT-1 is a simple and low cost 2-D EIT system. Its accuracy and operation are tested in different conditions. The system is designed to be upgraded to function as a multi-current generator adaptive system. Also by modification of the sampling circuit, SUT-1 will be able to detect the imaginary part of the signal can be detected. For this purpose voltage sampling has to be carried out during zero-crossing instead of peak sampling. The system was tested under in-vitro conditions. In order to perform in-vivo measurement, the IEC-601 safety standard has to be observed. The system needs some changes using an isolation component, e.g. opto-couplers in the data acquisition circuit, which can provide a complete electrical isolation. It is believed that the SUT-1 can be also used for different EIT applications such as industrial process control.
The different hardware parts of an engineered EIT system namely SUT-1 were investigated. SUT-1 also has its own limitations in practical use. Primary studies are under way to increase the SUT-1's capabilities. This can be achieved through the use of better electrodes (e.g. active electrodes), faster data acquisition technique, multi-frequency and real time 3-D image processing. Industrial applications of system for optimizing the metallurgical and chemical processes are some of the other potential applications of SUT-1. The idea of this paper to make available a system design for a simple EIT system and has no claim that this is a state of the art EIT system. For a good reference to EIT hardware we refer to a new book edited by David Holder . The book contains a good overview about the design of EIT instrumentation.
- Grimnes S, Martinsen OG: "Bioimpedance and Bioelectricity Basic". Academic Press; 2000.Google Scholar
- Webster JG: "Electrical Impedance Tomography". Adam Hilger 1990.Google Scholar
- Barber DC, Brown BH: "Imaging Spatial Distributions of Resistivity using Applied potential Tomography". Elect Letters 1983.,19(22):Google Scholar
- Mc Adams ET, Jossinet J: "Tissue impedance: a historical overview". Physiol Meas 1995, 16.Google Scholar
- Cheney M, Isaacson D, Newell JC: "Electrical Impedance Tomography". SIAM Review 1999.,41(1):Google Scholar
- Kaipio JP: "Simulation of the Heterogeneity of Environments by Finite Element Methods". University of Kuopio, Department of Applied Physics, Finland Report 1994., (4/94):Google Scholar
- Mu Z, Wexler A: "Electrical Impedance Computed Tomography Algorithms And Application". University of Mantioba, Department of Electrical and Computer Engineering, IEEE Proc 1994.Google Scholar
- Savolainen T: "An EIT Measurement System for Experimental Use". University of Kuopio, Department of Applied Physics, Finland, Report 1996., (2/96):Google Scholar
- Savolainen T: "Designing of a modular adaptive EIT measurement system". Department of Applied Physics, Univ. of Kuopio, Finland 1999.Google Scholar
- Movafeghi A, Soleimani M, Nateghi A, Mireshghi SA: "Introducing SUT-1, A Simple and Efficient EIT System". XI International Conference on Electrical Bio-Impedance Oslo, Norway 2001.Google Scholar
- Soleimani M, Movafeghi A: "Image Reconstruction Methods for Electrical Impedance Tomography (EIT) on SUT-1 system". 23rd Annual International Conference of the IEEE Engineering in Medicine and Biology Society, Istanbul, Turkey 2001.Google Scholar
- Metherall P: "3-D EIT of the human thorax". Ph D thesis, University of Sheffield 1998.Google Scholar
- User's Manual: "PCL 812 PG, Enhanced multi-lab Card". Advantech Co. Ltd Google Scholar
- Blad B, Johnnesson J, Johnsson G, Bachman B, Lindsrom K: "Waveform generator for electrical impedance tomography(EIT) using linear interpolation with multiplying D/A converters". Journal of Medical Engineering & Technology 1994.,18(5):Google Scholar
- Koukourlis CS, Kyriacou GA, Sahalos JN: "A 32-electrode data collection system for Electrical Impedance Tomography". IEEE Transactions on Biomedical Engineering 1995.,42(6):Google Scholar
- Eyuboglu BM: "Distinguishability analysis of an induced current EIT system using discrete coils". Phy Med Biol 2000.,45(7):Google Scholar
- Faes TJ: "The electrical resistivity of human tissue (100 Hz-10 MHz): a meta-analysis of review studies". Physiol Meas 1999., 20: Google Scholar
- Martinsen OG, Grimnes S, Schwan HP: "Interface Phenomena and Dielectric Properties of Biological Tissue". Encyclopedia of Surface and collied Science, Marcel Dekker Inc 2002.Google Scholar
- Holder DS: Electrical Impedance Tomography'. IoP Publishing, Bristol, UK 2004.Google Scholar
This article is published under license to BioMed Central Ltd. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/2.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.