IJST, Transactions of Electrical Engineering, Vol. 38, No. E1, pp 1-20

Printed in The Islamic Republic of Iran, 2014 © Shiraz University

DESIGN AND IMPLEMENTATION OF PRECISE HARDWARE FOR ELECTRICAL IMPEDANCE TOMOGRAPHY (EIT)*

M. KHALIGHI1** , B. VOSOUGHI VAHDAT2, M. MORTAZAVI3 AND M. MIKAEILI4 1School of Engineering and Science, Sharif University of Technology, International Campus, I. R. of Iran

Email: khalighi58@yahoo.com 2School of Engineering and Science, Sharif University of Technology, Tehran, I. R. of Iran

3School of Engineering, Islamic Azad University-Abhar Branch, I. R. of Iran 4Dept. of Engineering, Biomedical Engineering Group, Shahed University, Tehran, I. R. of Iran

Abstract– Electrical Impedance Tomography (EIT), is one of the safest medical imaging technologies and can be used in industrial process monitoring. In this method, image of electrical conductivity (or electrical impedance) distribution of the inner part of a conductive subject can be reconstructed. The image reconstruction process is done by injecting an accurate current into the boundary of a volume conductor (Ω), measuring voltages around the boundary (∂Ω) and transmitting them to a computer, and processing on acquired data with software (e.g. MATLAB). The image would be reconstructed from the measured peripheral data by using an iterative algorithm. A precise instrumentation (EIT hardware) plays a very important and vital role in the quality of reconstructed images. In this paper, we have proposed a practical design of a low-cost precise EIT hardware including, a high output impedance VCCS (Voltage-Controlled Current Source) with pulse generation part, precise voltage demodulator and measuring parts, a high performance multiplexer module, and a control unit. All the parts have been practically and accurately tested with successful results, and finally the proposed design was assembled on PCB. The quality of experimental results at the end of this paper, (reconstructed images by using the implemented system), confirms the accuracy of the proposed EIT hardware.

Keywords– EIT, electrical impedance tomography, EIT hardware, EIT instrumentation, EIT current source

1. INTRODUCTION

Electrical Impedance Tomography (EIT) is a relatively new imaging technique. In this method, an image

of the inner part of a conductive domain (Ω) can be made with an array of external electrodes which are

located on the boundary of domain (∂Ω). In this imaging method, the image of electrical conductivity (or

impedance) distribution of the internal part of a typical conductive subject can be reconstructed [1]. EIT

procedure includes injecting an accurate current into the boundary of domain (∂Ω) via a pair of electrodes,

measuring the boundary voltages by means of other electrodes around the boundary and transmitting them

to a computer, and at the end, processing the acquired data with software (e.g. MATLAB) to reconstruct

the image. Human body tissues contain a wide range of conductivities, and hence the potential exists to

use EIT to carry out medical imaging using the conductivity as the parameter to be mapped [2]. As a matter of fact, EIT is a challenging problem. This technique has some advantages compared to

other methods, including: simplicity of application, no hazard to the patient (such as X-ray), low cost and portable, and the high speed of data collection and image reconstruction [3]. Although EIT systems suffer

Received by the editors January 21, 2013; Accepted May 18, 2014. Corresponding author

IJST, Transac

2

from poor imof unknownquality [4]. algorithms tmost imporReconstructalgorithm [7

This padesign is illthat, some eend, final re

In general, E

EIT-sensors

iterative alg

data acquisi

The precisio

proposed EI

different pa

1- VCCS a VCC

2- Multip3- Contro4- Voltag

The co

(∂Ω) via th

types of pu

waveform.

demodulatio

the electrod

block. Cont

communica

ctions of Electr

mage resolutn skin–electrEIT still hasto reconstrucrtant of themted image q7]. aper is focuslustrated andexperiments econstructed

EIT instrum

s and a PC.

gorithm. In o

ition part (to

on of all uni

IT system is

arts:

(Voltage-CoC (Voltage-toplexer modulol unit. ge measurem

onstant and a

he multiplexe

ulse train sim

They can be

on portion. I

des voltage, “

trol unit is ap

ating with the

rical Engineeri

tion, the tranrode impedans technical dct the image

m is hardwarquality main

sed on hardwd explained, done by usiimage of ou

Fig. 1. The ma

2. TO

entation com

The image i

ur design, th

o measure vo

its plays a ve

illustrated in

ontrolled Cuo-current cone.

ent part cont

accurate curr

er part (MU

multaneous w

e used for sa

In order to m

“V-Demo.” a

pplied for co

e PC for data

M.

ing, Volume 38

nsfer impedannce on measudifficulty in ts. However, re improvem

nly depends

ware implemand each pa

ing the impleur EIT system

ain block diag

OP VIEW O

mprises digita

is finally rec

he analog sec

oltages with

ery importan

n Fig. 1. The

urrent Sourcenverter) part,

taining a volt

rent produce

UX). The sec

with maximu

ampling of re

measure the

and “V-Mete

ontrolling the

a transaction.

Khalighi et al.

8, Number E1

nces must sturement erroterms of devthere are so

ments to acqon the bou

mentation of Eart of the desemented syst

m will be com

gram of propo

OF PROPOS

al and analog

constructed f

ction of hardw

enough prec

nt role in rec

e block diagr

e) which incland a pulse

tage demodu

ed by the VC

cond task of

um point of

eal and imag

boundary vo

er” blocks sh

e multiplexer

. The assemb

l.

ill be measurors can also hveloping hardome methodsquire a high-undary data

EIT. First ofsign will be tem and thei

mpared with

sed EIT hardw

SED DESIG

g circuits, an

from the acq

ware include

cision). The

onstructed im

ram of the pr

udes a VCOgenerator.

ulator and a v

CCS is injec

f the current

positive pea

ginary section

oltage, mean

hould be app

r module an

bled EIT hard

red with highhave a signifdware for das to improve-accuracy me

accuracy an

f all, the top described in

ir results wilother works.

ware design

GN

n array of ele

quired data (

es an accurat

digital part

mage quality

roposed EIT

O as a wavefo

voltmeter.

cted into the

source sect

aks and zero

ns of measur

ning demodu

plied respecti

d measurem

dware is sho

h accuracy. Tficant effect

ata acquisitioe imaging queasured datund the recon

view of our n detail as wll be presente.

ectrodes whi

voltages) by

te current sou

contains con

y. The top vi

hardware (F

orm generato

e boundary o

ion is gener

points of th

red voltage i

ulating and m

ively after m

ent parts, an

wn in Fig. 2

June 2014

The effect on image

on and the uality. The um [5, 6]. nstruction

proposed well. After ed. At the

ich acts as

y using an

urce and a

ntrol unit.

iew of the

Fig. 1) has

or, a filter,

of domain

rating two

he current

in voltage

measuring

multiplexer

nd also for

.

June 2014

Electrical Imfrequency rcurrent is topositive andVCCS usedparts. Desigprecision, wsystems in mdigital or anoutput waveideal voltagaddition, thimpedance, range of losource can a

The cu10KHz andmust be in tsatisfy the a(Voltage-Cofilters in ser

a) Wavefor

Schema(EXAR Inc

mpedance Trange and wito use a voltad negative fed in the EIT sgn of the VCwhich means most cases inalog form [eform (i.e. exge source), whe main chara

linearity in oad, and propameliorate th

urrent source d 1MHz and the range of above conditontrolled Osrial connectio

rm generatio

atic circuit o.) has been u

Design

3. VOLTA

omography th a large vaage-controlleeedback arousystems, conC in EIT systhat it must

is the sinusoi[9]. The mainxact sinusoid

wide operatioacteristics foconverting

per workinghe image quain EIT systebe able to s

more than 10tions. Figure

scillator) parton), and a VC

Fig

ion

of the wavefused in the w

n and implemen

I

Fig. 2. Ass

AGE-CONTR

(EIT) systemariation in loaed current sound a high gnsists of wavstems is espehave high ouidal waveforn features fodal waveformonal bandwidor the VCC pvoltage to c

g in a broadality to a certems must be support the 00KΩ as wee 3 shows tht, a ButterwoCC (Voltage

g. 3. Flowchar

form generatwaveform os

ntation of preci

IJST, Transacti

sembled EIT h

ROLLED C

ms require acad impedancource (VCCSain operation

veform generecially imporutput impedarm which is or a typical wm without anydth, and steapart to have

current, precd range of frtain extent [1able to delivload betwee

ell [12], therehe flowchart orth band-pae-to-Current

rt of proposed

tion part is sscillator part

ise hardware fo

tions of Electric

hardware

CURRENT S

ccurate currece. The simpS), which canal or instrumrator and Vortant. The VCance. Generaproduced by

waveform gey jitter), a lo

ady amplitudan excellen

ision of outprequency [1011]. ver the currenen 100Ω andefore an exceof the propo

ass filter (BuConverter) p

VCCS design

shown in Figt as a VCO.

for…

cal Engineering

SOURCE

ent sources tlest techniqu

an be definedmentation amltage-to-CurrCC part requally, the wavy the wavefoenerator are aw output imp

de over all thnt VCCS are put wavefor0]. The high

nt over a freqd 10KΩ [9]. ellent currentosed VCCS. utterworth lopart.

n

g. 4. As it cAn AD844

g, Volume 38, N

that work ovue to obtain ad as a combmplifier [8]. rent Convert

uires stabilityeform used i

orm generatias follow; anpedance (sim

he frequencybased on him, supportin

h-performanc

quency rangeIts output im

t source desiIt consists o

ow-pass and

can be seen, IC has been

Number E1

3

ver a wide a constant ination of A typical

ter (VCC) y and high in the EIT on part in n accurate milar to an range. In igh output ng a wide ce current

e between mpedance gn should of a VCO high-pass

XR-2206 placed at

IJST, Transac

4

the output ovalue at thAD844, is a

b) Butterw

In ordeoutput impe250KHz, haoperational design the fcircuit of thpass filter (filter (U4 an

ctions of Electr

of the VCO,e output of

about 690Ω a

worth band-p

er to have anedance for thas been desifrequency r

filter for othehe filter is sh(U1 and U2) nd U5) with a

Fig. 5. Bu

rical Engineeri

, to amplify f waveform and with this

Fig. 4. Voltag

pass filter

n exact wavhe next stageigned and plrange of the er frequencieown in Fig. with cutoff

a cutoff frequ

utterworth Ba

M.

ing, Volume 38

the sinusoidgeneration p

s IC, is about

ge-controlled

veform withoe (VCC), a Blaced betweedesigned EI

es, the Butter5. As it is shfrequency o

uency of 10 K

and-pass filter

Khalighi et al.

8, Number E1

dal waveformpart (measurt 14Ω).

oscillator (wa

out any jitterButterworth en the VCO IT system wrworth coeff

hown, the filtf 250 KHz aKHz that are

r which is put

l.

m and also tred output i

aveform gener

r, noise and band-pass fiand VCC p

ould be betwficients have ter consists oand also a fo connected s

between the V

to create a limpedance o

ration part)

distortion, ailter Rangingparts of curreween 10 and

been cited iof a fourth-oourth-order Bserially.

VCO and VCC

ow output imof the circui

and also to cg between 10ent source. H

d 250KHz. Inn [13]. The srder ButterwButterworth

C parts

June 2014

mpedance it without

create low 0KHz and Hence the n order to schematic

worth low-high-pass

June 2014

c) Voltage-

The enwere done configuratio VCCS VCCS VCCS VCCS

These

values suchAccording tmodified. Toperational schematic cPSPICE), acomponentsTOA1 circu

Here, if:

We can get:

The ou

7, V1 represmeasured ooptional valan oscillosc

-to-current c

nd structure oto improve

ons of a VCCS based on AS based on DS based on TS based on T

experimentsh as differentto results ofThe main se

band-widthcircuits of d

and their resus value, accouit, (2) should

:

Fig. 6. P

utput impedasents the me

output voltaglue [2]. In orcope, the inpu

Design

converter

of mentionede and modiCS:

AH (AdvanceDOA (Double

OA1 (TripleOA2 (Triple

s were done t resistance vf the primaryelected criteh, and the mdifferent VCults have beeording to thed be satisfied

IL =

Proposed struc

ance Zout of aeasured outpge of R1, whrder to measuut impedance

n and implemen

I

d flowchart ify its opera

ed Howland)e-Operationae-Operationae-Operationa

with identicvalues, diffey practical teeria in the pminimum anCSs which

en shown in e best resultd and therefo

=

–

IL

cture of the VC

a typical currput voltage oen S is closeure the outpue of the voltm

ntation of preci

IJST, Transacti

is Voltage-toation. Sever

, [14, 15]. al Amplifier)l Amplifier fl Amplifier f

cal VCOs, drent types oests, the besprimary testsnd maximumhave been uAppendix A

t in experimeore IL is calcu

= 0 , or

.

.Vi

CC section, w

rent source cof R1, when ed. RP and Rut impedancemeter (oscillo

–∙

ise hardware fo

tions of Electric

o-Current conral essential

, [16]. form 1), [17]form 2), [9].

different VCCf OP-AMP, t structure os were basem loads thaused in prac

A. Proposed sents is showulated by (3)

R1 . R4 =

with suggested

can be calculthe switch S

R1 are connee and also loaoscope) shou

for…

cal Engineering

nverter (VCCtests were

].

C structures,and also dif

of the VCC ped on the ouat each VCCctical tests astructure of

wn in Fig. 6.).

= R2 . R3

components v

lated by (4). S is opened ected at the ad voltage ofuld be consid

g, Volume 38, N

C), many expdone on f

, different cofferent rangepart was selutput impedCS can suppand simulatithe VCC par For VCCS

values

As it is showand V2 repreoutput of VCf a typical Vdered.

Number E1

5

periments four main

omponent es of load. lected and dance, the port. The ons (with rt with its based on

(1)

(2)

(3)

wn in Fig. esents the CCS with

VCCS with

(4)

IJST, Transac

6

Figure to the circuwhich the Vbe exact sinduring the tImpedance particular frin Fig. 8A. resistor, wh

Fig. 8. (A) ThVCCS, in dioutput voltagwith PSPICE

ctions of Electr

8A shows thuit of Fig. 6. VCCS can sunusoid and ttest. The ampof a current

requency is aThe best va

hich creates h

he curve reprefferent frequege of implemeE in different f

rical Engineeri

Fig. 7. A m

he results of In the curve

upport in a spthe relationshplitude of lot source depalso restrictelue for R1 to

higher output

esents the ranencies and difented VCCS ifrequencies

M.

ing, Volume 38

method for mea

f other practice, Maximumpecific frequhip of the in

oad current wends on inved by the valu

o R4 was selet impedance

nges of maximfferent values in different fre

Khalighi et al.

8, Number E1

asuring VCCS

cal tests to d

m allowable luency. The onput voltage was adjusted erse of frequue of output ected as 1KΩcompared to

mum allowableof R1 to R4 a

equencies. (C

l.

S output imped

determine theload represenoutput wavefo

of VCC paron 1mA in

uency [9]. Timpedance. Ω. This is duo others.

e load which caccording to t) Output Impe

dance [2]

e best value nts the maximform of VCCrt and load call frequenci

The maximumAs a result a

ue to the VC

can be supporthe circuit inedance of pro

for R1 to R4 mum value o

CS (load voltcurrent musties. In fact, tm allowable according to C structure w

rted by the imFig. 6. (B) O

oposed VCCS,

June 2014

according of load in age) must t be linear the output load in a the curve

with 1KΩ

mplemented pen circuit , simulated

June 2014

With twhich decrewas tested AMPs werethis part thecircuit (openof simulated d) Pulse ge

A pulstypes of syncurrent wademodulatindiagram. Thof the pulsepulse widthpulse train voltage in th

The data acvoltage acq

this resistanceases to abouwith resistor

e heated up. Te output voltn circuit voltd VCCS in P

enerator

e generationnchronous puaveform (peng the boundhe input of thes with R2. Ah can be variand also thehe demodula

Fig.

cquisition syquiring parts

Design

ce value the ut 8KΩ in 2rs less than The result oftage of the Vtage), in ope

PSPICE.

n module hasulse train wiak detectiondary voltageshe pulse gene

As shown in Fied by increae signal of Pator section.

F

10. The outpu

4

stem can lims of the EI

n and implemen

I

maximum a250KHz as sh

1KΩ (~0.5Kf the other prVCCS was merational freq

s been attachith the maximn and zeros in order to erator is connFig. 10, sectiasing C1 fromoint B may

Fig. 9. Pulse ge

ut generated pu

4. VOLTAG

mit the overaIT system. R

ntation of preci

IJST, Transacti

allowable loahown in the KΩ), howevractical test fmeasured whquency range

hed to the Vmum point o detection measure the

nected to theions A and Bm 1nF to 10be used for

eneration sche

ulses. (A) Peak

GE MEASUR

all accuracy Related to

ise hardware fo

tions of Electric

ad in frequeupper curve

ver, the wavefor the propohen there wae. Part C of F

VCCS circuiof positive peof the curr

em. Figure 9e output of thB depict the wnF. For instasampling an

ematic circuit

k detection (B)

REMENT

[18]. Errors the topology

for…

cal Engineering

encies less the of Fig. 8Aeform was d

osed VCCS isasn’t any loaFig. 8 shows

t design in oeaks and zerrent).The pu

9 shows the phe filter. We worthwhile uance, by regnd extracting

) Zero detectio

arise from by that is ap

g, Volume 38, N

han 100KHz. The circuitdistorted ands shown in F

ad at the outpthe output im

order to genro points of tulses were pulse generatmay shift th

usage of thisgulating R2, tg real part o

on

both current pplied for m

Number E1

7

z is 12KΩ t in Fig. 6 d the OP-Fig. 8B. In put of the mpedance

nerate two the output used for

tor circuit he position s part. The the output f the load

drive and measuring

IJST, Transac

8

(measuring addition to are connecte a) Demodu

In ordestage, the vofor measureshould be tpositive peato the 90 dknown as p[19].

As it ibuffers are entered via part of the lport D mustworks as thamplifier) athose menti[20]. To mepulse should

ctions of Electr

pattern) in eavoiding meed to the cur

ulation of ele

er to measureoltage demoement. As antaken from tak. On the mdegree phasepulse-sample

is shown in entered to thports C and load voltaget be connectehe peak deteare the demoioned voltageeasure the imd be applied

R7 = R

rical Engineeri

each momeneasurement errent source i

ectrodes volt

e the boundadulator sampn example, ithe load volt

mentioned po shift. The m demodulatio

Fig. 11, bohe high speeD, come fro

e, port C mued to the maxector). Hencodulated eleces. The gain maginary paas well.

R8

Fig. 11. Prop

M.

ing, Volume 38

nt, only one rror becauseis not measur

ltage

ary voltage, aples the voltain order to mtage when th

oints, the volmethod of mon. The sugg

th electrode ed CMOS-loom the pulsest be connecximum pointe the input trode voltagvalue determ

art of the loa

osed design fo

Khalighi et al.

8, Number E1

voltage can e of commonred.

at first the elage waveformmeasure the he injecting ltage of capamodulation igested meth

voltages afogic analog me generator pcted to the pt detector (tovoltages of es and the omined by R7

ad voltage th

Gain

for demodulati

l.

be measuredn mode effect

ectrodes voltm in specificreal portioncurrent wou

acitance coms based on od does not

fter passing multiplexers

part of the cuositive peak

o the output oAD625 (proutput voltag, R8 and R9 i

he pulse of p

n = +1

ing of electrod

d via two pats, the voltag

tage must bec points and pn of the loaduld be on th

mponents of tsample-and-require mul

through the s (74HC4053urrent source

detector (poof the pulse gogrammable e of the IC is optional anpoint B and

de voltages

articular elecge of electrod

e demodulateprepares sam

d voltage, thehe maximumhe subject ishold approaltiplier or ou

high pass f3). The contre. To measuroint B of Figgenerator par

gain instrumis differentiand is calculatzero detecti

June 2014

ctrodes. In des which

ed. At this mpled data e samples

m value of s zero due ach that is utput filter

filters and rol pulses re the real g. 11) and rt, when it mentation

al value of ted by (5) ion output

(5)

June 2014

b) Voltage

At this

value. The

measuremen

ADCs by i

conversion

between 13

analog volt

conversion

whenever it

mentioned A

A high-spee

As it is see

used as the

For a 3

are applied

The multipl

32-electrode

pin can be u

measureme

s stage the de

digital valu

nt part, AD1

itself, but its

time (i.e. req

to 260μs w

tage signal

or sampling

t is required

ADC can sen

ed multiplex

en in Fig. 12

analog multi

32-electrode

for injection

lexer is appli

e system, 5 a

used. Each la

Design

ent

emodulated v

ue is then tr

1674 with 12

s ADCs can

quired time f

which is not

at a specifi

time is exac

d. This is du

nse 2.44 mV

xer module w

2, the schem

iplexer in ou

EIT system

n and sink po

ied for sharin

address lines

atch is applie

Fig. 12.

n and implemen

I

voltage is me

ransmitted t

2-bit resolut

nnot be used

for conversio

an exact am

ic point in

ctly 10μs [22

e to the 12-

changing on

5. MU

with 32 outp

matic circuit o

ur design.

m, eight ICs o

orts of curren

ng the curren

s are needed;

ed for a pair o

. The proposed

ntation of preci

IJST, Transacti

easured, hen

to the micro

tion is applie

d in measure

on of an ana

mount of tim

the range o

2], hence it is

bit resolutio

n the analog

ULTIPLEX

puts can be a

of the multip

of the menti

nt source and

nt source and

; hence four

of ADG506A

d design for m

ise hardware fo

tions of Electric

nce it must b

ocontroller (

ed as an AD

ement. Acco

alog value to

me. So it wou

of micro sec

s very interes

on of the IC,

voltage (one

XER

added to the

plexer modu

ioned multip

d the others a

d voltmeter b

address line

A (for lower

multiplexer mo

for…

cal Engineering

e converted

(ATmega128

DC. ATmega

ording to the

valid digital

uld not be p

cond. Howe

sting to captu

, for amplitu

e bit is used f

proposed in

ule is illustra

plexer are req

are used for

between mul

s of ADG50

and higher th

odule

g, Volume 38, N

from analog

8) of contro

a128 has a n

e data-sheet

l data) of its

ossible to ca

ver, in AD

ure the analo

ude between

for sign).

nstrumentatio

ated. ADG50

quired. Four

voltage mea

ltiple electro

06AKN plus

han 16 electr

Number E1

9

g to digital

l unit. In

number of

[21], the

ADCs, is

apture the

1674, the

og voltage

±5V, the

on design.

06AKN is

of which

asurement.

des. For a

its enable

rodes).

IJST, Transac

10

The con

Contro Comm Conne Conne Comm

As it c

output portsthe next stag

a) Data tran

Serial

therefore co

[23]. After

register mu

(ATmega12

voltage and

COM port.

microcontro

rates such a

ctions of Electr

ntrol unit conolling the mu

municating wecting to LCDecting to ADmunicating w

can be seen s should be bges.

nsaction via

communicat

onnecting the

completion

ust be trans

28) is conne

d vice versa.

. In this de

oller frequen

as 2.4, 9.6 an

rical Engineeri

ntains the foultiplexer pa

with a PC andD (if require

DC and controwith AVR pro

in Fig. 13, Abuffered to a

Fig

RS-232

tion is one o

e computer a

of the volta

smitted to

cted to the

. Hence the

esign based

ncy is assigne

d 115.2 kbps

M.

ing, Volume 38

6. CO

llowing taskart to determid matching wed), to displayolling that asogrammer fo

ATmega128amplify the p

g. 13. The prop

of the proto

and microco

age measurem

the comput

computer vi

data transac

on the stan

ed on 14.745

s, the data tra

Khalighi et al.

8, Number E1

ONTROL UN

ks: ine the state owith MATLAy the voltages a slave withor programm

is applied aport output cu

posed design o

ocols that are

ontroller, RS-

ment, the m

ter. In the

ia MAX232,

ction is done

ndard freque

56MHz. In th

ansaction err

l.

NIT

of each electAB for data tres and also nh handshakining via ISP m

as the microurrent and al

of control uni

e supported

-232 commu

measured volt

proposed E

, which conv

e by RS-232

encies and

he mentioned

ror is zero.

trode on eachransaction viumber of me

ng and interrumode.

ocontroller oflso to preven

t

by many ty

unication pro

tage stored i

EIT system,

verts the TT

2 serial com

baud-rates

d frequency

h moment. ia RS-232. easurements.upting signa

f control unint the loading

ypes of com

otocol is ofte

in the micro

, the micro

TL voltage t

mmunication

(bps) table

for all stand

June 2014

. ls.

it. All the g effect of

mputer and

en applied

ocontroller

ocontroller

o RS-232

via 9-pin

[23], the

dard baud-

June 2014

To show th

stimulation

cross pattern

as the stimu

In the

electrodes a

applied thro

procedure w

pattern, it is

the number

pattern as th

in the outer

and decreas

Fig.

To evaluate

cylindrical p

will be show

In all exper

water. A cu

electrodes.

experiments

cylindrical p a) Experim

In the

of a 16-elec

image of su

shown in F

illustrated in

parts C and

7

he experime

(current inje

n. In this pap

ulation and m

adjacent (ne

and the volta

ough the nex

will be contin

s common to

r of electrod

he system sti

r region of th

ses rapidly as

. 14. Adjacent

e performanc

phantom wit

wn and desc

riments, the s

urrent signal

Adjacent p

s. All imag

phantom is m

ment 1

first experim

ctrode cylind

ubjects havi

ig. 15. Parts

n parts C and

E the image

Design

7. STIMULA

ental results

ection) and

per the adjac

measurement

eighboring) p

ages are mea

xt pair of elec

nued until ea

o use all N ×

es. ATmega

imulation an

he imaged o

s a function o

t pattern as the

ce of the sys

th a few test

cribed, and so

saline used i

(~1mA) wit

pattern is u

ges are reco

made of Plex

ment two piec

drical phanto

ng less cond

s B, E and F

d D. In parts

e is reconstru

n and implemen

I

ATION AND

of the prop

voltage mea

cent pattern t

topologies.

pattern, show

sured from s

ctrodes, and

ach possible

× (N − 3) me

a128 in the c

nd measurem

bject. The cu

of distance.

e stimulation a

8. EXPERIM

stem to recon

t objects. Th

ome element

in phantoms

th 20 KHz fr

sed as the

onstructed b

xiglas with a

ces of a plas

om. The aim

ductivity tha

F show the fi

s B, D, and F

ucted onto ele

ntation of preci

IJST, Transacti

D MEASUR

posed EIT s

asurement, su

that is the mo

wn in Fig. 14

successive pa

the voltage m

pair of adjac

easurements

control unit

ment protocol

urrent densit

and measurem

MENTAL R

nstruct imag

he images rec

ts that can af

is prepared w

frequency is

stimulation

by using the

radius of 15c

stic (Teflon)

m of experim

an saline. Th

fine-model im

F the image o

ements.

ise hardware fo

tions of Electric

REMENT PA

system, diffe

uch as: adjac

ost common

4, the curren

airs of adjac

measuremen

cent electrod

in most reco

has been pr

s. In this pat

ty is highest

ment protocols

RESULTS

es, some exp

constructed

ffect the qua

with solution

injected to th

and voltag

e Eidors M

cm.

shaft have b

ment is to eva

he reconstru

mages of pha

of phantom, i

for…

cal Engineering

ATTERNS

erent topolog

cent pattern,

pair-drive pr

nt is applied

ent electrode

nts will be rep

des is used to

onstruction a

rogrammed b

ttern the curr

t between the

s of the propos

periments w

by using the

ality of imag

n of NaCl an

he boundary

ge measurem

MATLAB pa

been placed i

aluate the qu

ucted images

antom, and t

is reconstruc

g, Volume 38, N

gies can be

, opposite pa

rotocol [24]

through two

es. The curre

peated for ot

o inject curre

algorithms, w

based on the

rent is injecte

e injecting e

sed EIT system

were carried o

e implemente

ges will be in

nd de-ionized

y of phantom

ment protoc

ackage [25,

in front of el

uality of reco

s of this pha

the coarse m

cted onto nod

Number E1

11

used for

attern and

is applied

o adjacent

ent is then

thers. The

ent. In this

where N is

e adjacent

ed mainly

electrodes,

m

out on the

ed system

ntroduced.

d distilled

ms via two

col in all

26]. The

lectrode 1

onstructed

antom are

models are

des, but in

IJST, Transac

12

Fig. 15. (A) reconstructedelements. (F) b) Experim

In this 4 and 14 ofthe quality Fig. 16 illusand coarse m

Fig. 16. (A) reconstructedelements. (F) c) Experim

In this cylindrical subjects in and coarse m

ctions of Electr

Phantom withd onto eleme) Fine model r

ent 2

experiment,f a 16-electroof reconstrustrate the finmesh genera

Phantom withd onto eleme) Fine model r

ment 3

experimentphantom. Thdifferent conmodels are sh

rical Engineeri

h two pieces ents. (D) Coareconstructed

two pieces oode cylindricucted image onal reconstruation models.

h two pieces oents. (D) Coareconstructed

, four pieceshe experimenductivities hown in part

M.

ing, Volume 38

of plastic shaarse model ronto nodes

of a metalliccal phantom of the subjec

ucted images

of metallic sharse model ronto nodes

s of the metnt is conduclocating in

ts B to F of F

Khalighi et al.

8, Number E1

aft. (B) Fine meconstructed

c (Aluminumwith 30 cm cts having asof the phan

haft. (B) Fine meconstructed

tallic and placted to evalucomposite seFig. 17.

l.

model reconstronto nodes.

m) shaft have diameter. Ths much cond

ntom (shown

model reconstonto nodes.

astic shafts huate the qualetting. The f

ructed onto n(E) Fine mo

been placedhe aim of thiductivity as s

in part A) b

tructed onto n(E) Fine mo

have been pulity of reconfinal reconst

nodes. (C) Coaodel reconstru

d in front of eis test is to usaline. Parts by using diff

nodes. (C) Coaodel reconstru

ut in the 16-nstructed imatructed imag

June 2014

arse model ucted onto

electrodes understand

B to F of ferent fine

arse model ucted onto

-electrode age of the ges in fine

June 2014

Fig. 17. (A)Coarse modreconstructed d) Experim

In the phantom. Tthorax-shapC (of Fig. 1

Fig. 18. (A) (C) Fine mod e) Experim

The aimimages, andparameter. Ihyper-paramparameter sobtain a uscauses the iaction. As parameter vin terms of p

) Phantom witdel reconstrucd onto elemen

ment 4

following, thhe goal of th

pe phantom. A8) the image

A 16-electroddel reconstruc

ment 5

m of conducd evaluate tIn fact hypemeter selectiselection shoseful reconsimage to be it can be se

values are shprecision and

Design

th four piecescted onto el

nts. (F) Fine m

hree plastic ahis test is evaAs it can be e is reconstru

de thorax-shapcted onto node

cting this expthe quality r-parameter on is to prod

ould producetruction [24dominated b

een in Fig. hown. Obviod quality.

n and implemen

I

s of metallic alements. (D)

model reconstru

and metallicaluation of qseen in Fig.

ucted onto no

pe phantom wes

periment is toof the fina(λ) is a scalduce the bes

e solutions th4]. Practicallby noise and19, few recusly the reco

ntation of preci

IJST, Transacti

and plastic shaCoarse mod

ucted onto no

subjects havquality of the

18B, the imodes.

with some subj

o show the el images relar that contrst reconstruchat preserve ly in image

d as it is incronstructed imonstructed im

ise hardware fo

tions of Electric

afts. (B) Fine del reconstru

odes

ve been place images reco

mage is recon

jects. (B) Fin

effects of hypeconstructed rols the amoction with hi

as much of reconstruct

reased, noisemages of thmage with λ

for…

cal Engineering

model reconsucted onto n

ced in a 16-eonstructed bystructed onto

e model recon

per-parametewith differ

unt of regulgh quality a

f the measuretion, insuffic is filtered th

he phantom = 0.01, repr

g, Volume 38, N

structed onto

nodes. (E) Fi

electrode thoy using a 16-o elements, b

nstructed onto

er (λ) in recoent values oarization. Th

and resolutioed data as pcient hyper-phrough the swith differe

resents the b

Number E1

13

nodes. (C) ine model

orax-shape -electrode but in part

o elements.

onstructed of hyper-he goal of n. Hyper-

possible to parameter smoothing ent hyper-best image

IJST, Transac

14

Fig. 19. SomEIT system. f) Experim

In the reconstructeframe acqui16-electrode

In expecylindrical pof voltage wnoise can breconstructeframes of vand F of Figof voltage fof part C whthe averagefinal recons

Fig. 20. Somfour pieces o(D) n=5. (E)

ctions of Electr

me images whiThe image wi

ment 6

current exed image quired by scane phantom eaeriment 6 sophantom as which meansbe seen on ed after scanvoltages. So g. 20 are the frames. The hich is recon

e of voltage structed imag

me reconstrucof metallic ann=10. (F) 3D

rical Engineeri

ich were reconith λ=0.01 rep

periment, efality will be

nning an N-each frame ofme metallic shown in Figs that the imathis image

nning ten timthe effects o3D images oimage in par

nstructed afteframes inste

ges would be

cted images ofd plastic shaf image of pha

M.

ing, Volume 38

nstructed withpresents the be

ffect of nume analyzed. Inlectrode pha

f voltage conand plastic sg. 20.A. Theage is recons(near electro

mes, which mof noise are iof the phantort F reconstrer scanning oead of one fre improved.

f one phantomfts. (B) n=1. (antom in part A

Khalighi et al.

8, Number E1

h different λ vest image in te

mber of von fact, in adj

antom is calcntains 208 vosubjects are le image in pastructed afterodes 4, 8, 9eans image rimproved inom of part Aructed after sonly one timeframe, in ima

m, by using th(C) 3D image A with n=10

l.

values (hyper-erms of precis

ltage frame djacent patterculated by Noltages. located in diart B is reconr one time sc

9, and 10). Hreconstructio

n the mentionA, which are scanning ten e. This resultage reconstr

he average of (3D surface p

-parameter) byion and qualit

or numberrn the numbe

N×(N – 3). H

ifferent positnstructed by canning, therHowever th

on is done usned image. Treconstructetimes is sm

t can be extraruction proce

f n voltage fraplot) of phant

y using the 16ty

r of scanniner of voltage

Hence after s

tions of a 16-using only o

refore some e image in

sing the averThe images id by differenoother than tacted so thatess, the qual

ames. (A) Phatom in part A

June 2014

6-electrode

ng in the es in each canning a

-electrode one frame effects of part E is age of ten in parts C nt number the image t, by using lity of the

antom with A with n=1.

June 2014

g) Experim

The obimage qualinear the bouthis test, thelectrodes. evaluation iwall of the phantoms, iwhereas in separated fr

Fig. 21. Recelectrodes to

The ne

reconstructeapplied as Adjacent pacentral partelectrodes, m

Fig. 22. Shomeans of var

ment 7

bjective of exity. To evaluundary, in thhree phantomThe dimensiis near the bphantoms is

in some placreconstructe

rom the boun

constructed im show their ac

ext position ed image of the stimulatattern, and its. It is sensmeasuremen

owing qualityrious numbers

Design

xperiment 7 iuate the imaghe central parms are applions of phan

boundary. Ass about 2~3 ces show thad images by

ndary of phan

mages of subjeccuracy in fina

is the centrf subjects pution and meit is very sesitive to dis

nt error and n

y of reconstrus of electrodes

n and implemen

I

is to show thge quality, thrt and the whlied with dintoms and ins it can be scm, but in t

at the dark py 32-electrodntoms.

ects which are al image recon

ral part of tut in the ceneasurement tnsitive to co

sturbances innoise as well

cted images os

ntation of preci

IJST, Transacti

he effects of nhree positionhole phantomfferent num

nternal salineeen in Fig. 2the reconstruparts (smudge phantom a

located near tnstruction qua

the phantomntral part is topologies ionductive pan the bound[24, 27].

of the subject

ise hardware fo

tions of Electric

number of elns of a phantm (both centr

mbers of elece and subjec21, the distaucted imagesges) are connas it can be s

the boundary ality

m. As is illuspoor. This in all imagearts near the

dary shape o

t, which is loc

for…

cal Engineering

lectrodes in ttom surface ral part and nctrodes whiccts are identiance betweens by 8-electrnected to theen, the dark

of phantoms w

strated in Fiis because the reconstructe boundary aof the object

cated in centr

g, Volume 38, N

the final recowould be co

near the bounch are 8, 1ical. First pon subjects to rode and 16-e boundary k parts are co

with various n

ig. 22, the qhe pattern wtions in thisand insensitit, the positio

ral part of pha

Number E1

15

onstructed onsidered: ndary). In 6 and 32

osition for the inner

-electrode of image, ompletely

numbers of

quality of which was s paper is ive to the on of the

antoms, by

IJST, Transac

16

The laelectrode phthe subjectsthe subjectsshaft. But itof the 16-einstead of th

Fig. 23. Redifferent posreconstructio

The resubjects locinstead of p

Hence the adjacenparts near tdecreased sphenomenoand 32-elecsensitivity. topologies,

Although thresolution) extensively boundary voregular med

In thisvisually witinclude an system desielectrode phobjects loca

As a mother, but thof Fig. 24 a

ctions of Electr

ast position ihantom is ves which are los, there is a dt cannot be slectrode phahe central pla

econstructed isitions of the on processes

econstructed cated near thplastic shaft!

the result thnt pattern, whthe boundaryspecifically

on is that in ctrode phanto

Therefore the position

he EIT has bthat still neeresearched i

oltage data adical imagings part, the quth quality of EIT system

igned by Gohhantom and ated in phantomatter of fache size and paccording to

rical Engineeri

is the wholeery crude, anocated in thedark smudgeseen in the reantom showsastic shaft it

images of thrphantom (cen

image of the boundary,

hat can be achhen the numy will be incwhen some 8-electrode poms), and it in image rof subjects i

9. VISU

been developed to be ovein clinical di

and poor spatg devices. uality of recimages prodproposed b

harian et al ifrom this pooms are madt, the size of

position of sutheir paper

M.

ing, Volume 38

e phantom. And it does note phantom. Be in the centreconstructeds the correctshows a very

ree phantoms ntral part and

he 32-electrobut it does n

heived from mber of electcreased, but subjects ha

phantom thecauses morereconstructiois important a

UAL COMPA

ped substantiercome to miagnosis [29tial resolutio

constructed iduced by usiny Bera and in [32]. In fa

oint of view tde from similf all phantomubjects in theinformation

Khalighi et al.

8, Number E1

As shown int have accep

But despite thral part of th

d image of 32t position ofy light smud

with differennear the boun

ode phantomnot show any

this experimtrodes of pha

its sensitiviave been loce distance bee current denon accuracyas well, even

ARISON ON

ally over recmake it a clin

, 30], but duon of image, t

image usingng other impNagaraju in

act in both sythose systemlar materials ms and test eir own phan

[31-32]. Th

l.

n Fig. 23, thptable accurahe bad qualithis image, as2-electrode pf the subject

dge.

nt numbers ondary), and th

m also exactything, even

ment is, for aantom increaity to the cencated near thetween electrnsity in the cy, related ton when using

N RESOLU

cent years, thnically applicue to poor Sithe EIT syste

g the proposeplemented EIn [31] and a ystems, imag

ms are similar(plastic and

objects are dntoms are exahe mapped im

he reconstruacy to show tty and showi the position

phantom. Thts located ne

f electrodes. hese positions

tly shows tha light smud

n EIT systemases, the sysntral part ofhe boundaryrodes is morcentral part, o measuremg 32-electrod

UTION

here are mancable imaginignal to Noisems have no

ed EIT systeIT systems. T

DSP basedge is reconstr to the propmetal) as we

different in cactly plotted mage of pha

ucted image the correct ping inexact pn of the centre reconstrucear the boun

Subjects are s were identic

he correct podge, in its ce

m that worksstem sensitivf the phantomy. The reasore than the owhich result

ment and stde system.

ny challengesng [28]. EITse Ratio (SN

ot yet been ac

em will be The other EITd multi-frequtructed by us

posed systemell. comparison win the middl

antoms and t

June 2014

of the 8-position of position of ral plastic ted image

ndary, but

located in

cal in three

osition of entral part

s based on vity to the m will be on of this others (16 ts in more timulation

s (such as has been

NR) of the ccepted as

compared T systems

uency EIT sing a 16-

m. The test

with each le column their inner

June 2014

objects (in mof real phancolumn shoobjects. Thquality or re32 electrodincreased if

Fig. 24. Vis(A) reconstruusing a DSP our EIT syste

In this papedetail. Systeshown. TheSimplicity oEIT researcthe practicaoutput impewasn’t posssystem with

middle columntoms and theows three 16he third coluesolution of r

des of the imf the number

sual comparisoucted image bbased multi-f

em

er a practicaem performae novelty ofof design ma

chers for desial point of vieedance of thsible to be mh 16 electro

Design

mn of Fig. 24eir container6-electrode pumn shows treconstructe

mplemented of electrode

on on quality by using EIT frequency EIT

al low-cost pance has beenf this designade it possibigning and imew to improvhe current someasured wodes compar

n and implemen

I

4), have beenrs applied in phantoms relthe reconstrud image seenEIT system.s increases.

of final imagsystem propo

T system propo

9. CO

precise design checked w

n is its precible to have amplementative the precisource, which

with oscilloscred to the o

ntation of preci

IJST, Transacti

n drawn withimage reconlated to diffucted imagen in Fig. 24.C. Of course,

ges made by aosed by Bera osed by Goha

ONCLUSIO

gn of EIT inwith a saline tision which a low-cost inon of EIT ha

sion of the prh was finallycope. Higherothers is the

ise hardware fo

tions of Electric

h the exact prnstruction. Asferent EIT sys made by tC is obtained the quality

applying threeand Nagaraju

arian et al in [3

ON

strumentatiotank and somwas achieve

nstrumentatioardware. Somroposed systey increased r quality of e best reaso

for…

cal Engineering

roportion coms it can be seystems contathree differe

d by using onand resolut

e different impu in [31]. (B) 32]. (C) recon

on was propome experimeed without aon which come hard worem such as trto more thathe reconstr

on to verify

g, Volume 38, N

mpared to dieen in Fig. 24aining salineent EIT systnly 16 electrotion of imag

plemented EIreconstructed

nstructed imag

osed and desental results hany complexould pave therk has been drying to max

an 5 MΩ suructed image

the accurac

Number E1

17

imensions 4, the first e and test tems. The odes from

ge will be

T systems. d image by ge by using

scribed in have been x circuits. e way for done from ximize the uch that it es by our cy of the

M. Khalighi et al.

IJST, Transactions of Electrical Engineering, Volume 38, Number E1 June 2014

18

implemented EIT hardware. Future research includes improvement of the current EIT hardware in terms of accuracy and precision to become appropriate for clinical applications and development of EIT application in industrial and medical fields.

REFERENCES

1. Horesh, L. (2006). Some novel aproches in modelling and image reconstruction for multi-frequency electrical

impedance tomography of the human brain. PhD thesis.

2. Denyer, C. W. L. (1996). Electronics for real-time and three-dimensional electrical impedance tomography. PhD

thesis.

3. Yan, W., Hong, S. & Chaoshi, R. (2006). Optimum design of electrode structure and parameters in electrical

impedance tomography. Physiol. Meas., Vol. 27, pp. 291-306.

4. Boone, K. G. & Holder, D. S. (1996). Current approaches to analogue instrumentation design in electrical

impedance tomography. Physiol. Meas.17(suppl 1A) A229–A247.

5. Murphy, S. C., Chin, R. K. Y. & York, T. A. (2008). Design of an impeller-mounted electrode array for EIT

imaging. Meas. Sci. Technol., Vol. 19, pp. 1-12.

6. Goharian, M., Soleimani, M., Jegatheesan, A., Chin, K. & Moran, G. R. (2008). A DSP based multi-frequency

3D electrical impedance tomography system. Ann. Biomed. Eng., Vol. 36, pp. 1594-1603.

7. Yorkey, T. J. (1986). Comparing reconstruction methods for electrical impedance tomography. PhD thesis,

University of Wisconsin at Madison, Madison, WI 53706.

8. Filho, B., Brown, B. H. & Wilson, A. J. (2000). A comparison of modified Howland circuits as current

generators with current mirror type circuits. Physiol.Meas., Vol. 21.

9. Holder, D. S. (2005). Electrical impedance tomography: methods, history and applications (Series in Medical

Physics and Biomedical Engineering). Institute of Physics Publishing Ltd.

10. Khalighi, M., Vosoughi Vahdat, B., Mortazavi, M. & Soleimani, M. (2012). Practical design of low-cost

instrumentation for industrial Electrical Impedance Tomography (EIT). IEEE I2MTC, Graz.

11. Zhao, D. (2011). High output-impedance current source for electrical impedance tomography. IEEE 4th

International Conference on Biomedical Engineering and Informatics (BMEI).

12. Denyer, C., Lidgey, F. & Zhu, Q. (1994). A high output impedance current source. Physiol. Meas., Vol. 15, pp.

A79-A82.

13. Kugelstadt, T. (2009). Active filter design techniques (chapter 20), Excerpted from OP AMPs for everyone.

literature number: SLOD006A, published by Texas Instruments, third edition.

14. Cusick, G., Holder, D. S., Birkett, A. & Boone, K. (1994). A system for impedance imaging epilepsy in

ambulatory human subjects. Innov. Tech. Biol. Med.15(suppl 1A) A33–A39.

15. Jossinet, J., Tourtel, G. & Jarry, R. (1994). Performance and operation of a set of wideband current generators

for EIT. Innov. Tech. Biol. Med. 15(suppl 1A) A40–A46.

16. Zhang, Y. L., Zhui, X., Chaoshi, R., Wei, W., Dechun, Z. & Huiquan, Z. (2010). Study of voltage controlled

current source in electrical impedance tomography system. IEEE, iCBBE, 4th Int. Conference.

17. Shuai, Z., Guizhi, X., Huanli, W. & Duyan, G. (2006). Multi-frequency EIT hardware system based on DSP.

IEEE EMBS, 28th Annual international conference, China.

18. Smith, R. W. M. (1990). Design of a real-time impedance imaging system for medical applications”, PhD

Thesis, University of Sheffield.

19. Soleimani, M. (1998). Designing and making the electrical impedance tomography systems. Sharif University of

Technology, Msc thesis.

20. Data-sheet, (2000). AD625, programmable gain instrumentation amplifier. Analog Devices Inc., USA.

21. Data-sheet, (2006). ATmega128, 8-bit AVR microcontroller with 128K bytes programmable flash. Atmel Inc.

June 2014

22. Data-she

23. Rahafroo

24. Graham,

lung ima

25. Vauhkon

package

26. Adler, A

Physiol.

27. Dickin,

Measure

28. Bayford

Biomedi

29. Brown,

Science,

30. Hoetink

blood vo

Transac

31. Bera, T

inhomog

32. Goharian

3D elect

33. Khalighi

current s

Fig. A1. (A) (C) VCCS bform 2 [9]

eet, (2000). AD

oz, A. (2009).

, B. M. (2007

aging. Univers

nen, M., Lion

for the EIDO

A. & Lionhea

Meas., Vol. 2

F. & Wang

ement Science

d, R. H. (200

ical Engineeri

B. H. (2001).

Measuremen

, A. E., Faes,

olume change

tion on Medic

T. K. & Na

geneities in 2D

n, M., Soleim

trical impedan

i, M., Vosou

source for EIT APPE

A

VCCS basedbased on Trip

Design

D1674, 12-bit

. AVR Microc

7). Enhanceme

sity of Ottawa

nheart, W. R.

ORS project to

art, W. R. B.

27.

, M. (1996).

e and Technolo

06). Bioimped

ing, Vol. 8, pp

Medical imp

nt Science and

T. J. C., Mar

s during the h

cal Imaging, V

agaraju, J. (2

D-electrical im

mani, M., Jega

nce tomograph

ghi Vahdat, B

T systems. IJR

NDIX A. SCAND THE RE

d on Advancedle-Operationa

n and implemen

I

t 100 kSPS A/D

ontrollers and

ents in Electri

a. PhD thesis,

B., Heikkinen

o reconstruct tw

(2006). Uses

Electrical re

ogy, Vol. 7, N

dance tomogr

p. 63-91.

pedance tomog

d Technology,

rcus, J. T., Ke

heart cycle wit

Vol. 21, Issue

2011). Resis

mpedance tom

atheesan, A., C

hy system. An

B. & Mortaz

RTET, Vol. 6,

CHEMATICESULTS OF

d Howland [1al Amplifier f

ntation of preci

IJST, Transacti

/D converter. A

d their applica

ical Impedanc

Degree in ele

n, L. M., Vau

wo dimension

and abuses o

esistance tom

No. 3, pp. 247-

raphy (Electr

graphy and pr

Vol. 12, No. 8

erkkamp, H. J.

th electrical im

6.

tivity imagin

mography. Else

Chin, K. & M

nals of Biome

zavi, M. (201

No. 2, DOI: 0

C CIRCUITSF PRIMARY

14, 15]. (B) Vform 1 [17]. (

ise hardware fo

tions of Electric

Analog Devic

ations. Nas pu

ce Tomograph

ectrical and co

hkonen, P. J.

nal EIT image

of EIDORS:

mography for

-260.

rical impedan

rocess impeda

8, pp. 991–99

. J. & Heethaa

mpedance usin

ng of a rec

evier, Measure

Moran, G. R. (

edical Enginee

1). Designing

01.IJRTET.06

S OF MENTY PRACTIC

VCCS based on(D) VCCS ba

for…

cal Engineering

ces Inc., USA.

ublishing.

hy (EIT) imag

omputing engi

& Kaipio, J.

s. Physiol. Me

an extensible

process app

nce tomograph

ance tomograp

6.

ar, R. M. (200

ng a linear spo

onfigurable p

ement, Vol. 44

2008). A DSP

ering, Vol. 36

g a practical

.02. 239.

TIONED VCCAL TESTS

n Double-Opeased on Triple

g, Volume 38, N

ge reconstruct

ineering.

P. (2001). A

eas., Vol. 22.

software bas

lications. IOP

hy). Annual

phy: a brief re

02). Imaging

ot-electrode a

phantom wit

4, pp. 518-526

P based multi

6, No. 9.

high output

CCSS

erational Ampe-Operational

Number E1

19

tion for 3D

MATLAB

se for EIT.

P Science,

Review of

eview. IOP

of thoracic

array. IEEE

th circular

6.

-frequency

impedance

plifier [16]. Amplifier

M. Khalighi et al.

IJST, Transactions of Electrical Engineering, Volume 38, Number E1 June 2014

20

Figure A1 shows different schematic circuits of the VCCS which have been used in practical tests. As it can be

seen in Table A1, the results of several experiments are shown briefly. These tests were done on the breadboard with

different resistance values in condition of 20KHz frequency and 1mA load current. It can be concluded (from Table

A1) that the VCCS based on TOA1 (fourth row) is the best choice to have an efficient voltage-controlled current

source. It can support a load in range of 10Ω to 10KΩ linearly and also has more than 5MΩ output impedance which

calculated with (4).

Table A1. The results of VCCS primary physical tests [33]

VCCS Type

R1 (KΩ)

R2 (KΩ)

R3 (KΩ)

R4 (KΩ)

R5 (KΩ)

Allowable Load (Ω)

Measured (Ω)

PSPICE (Ω)

AH 1 2 1 1 1 10 ~ 1K 100K 420K DOA 1 1 1 1 5 10 ~ 5K 3 M 5.3M TOA1 1 1 1 1 5 10 ~ 10K >5 M 7.8M TOA1 2 2 2 2 5 10 ~ 10K 1 M 3.9M TOA1 5 5 5 5 5 10 ~ 10K 500K 1.6M TOA1 10 10 10 10 5 10 ~ 10K 500K 770K TOA1 22 22 22 22 5 10 ~ 10K 150K 340K TOA1 27 27 27 27 5 10 ~ 10K 100K 280K TOA2 1 1 1 1 5 10 ~ 10K 600K 2.5M