Review Article Bioelectrical Impedance Methods for Noninvasive Health Monitoring: A Review · 2019. 7. 31. · Review Article Bioelectrical Impedance Methods for Noninvasive Health

Review ArticleBioelectrical Impedance Methods for Noninvasive HealthMonitoring A Review

Tushar Kanti Bera

Department of Computational Science and Engineering Yonsei University Seoul 120749 Republic of Korea

Correspondence should be addressed to Tushar Kanti Bera tkbera77gmailcom

Received 31 August 2013 Revised 26 November 2013 Accepted 26 November 2013 Published 17 June 2014

Academic Editor Norio Iriguchi

Copyright copy 2014 Tushar Kanti Bera This is an open access article distributed under the Creative Commons Attribution Licensewhich permits unrestricted use distribution and reproduction in any medium provided the original work is properly cited

Under the alternating electrical excitation biological tissues produce a complex electrical impedance which depends on tissuecomposition structures health status and applied signal frequency and hence the bioelectrical impedance methods can beutilized for noninvasive tissue characterization As the impedance responses of these tissue parameters vary with frequenciesof the applied signal the impedance analysis conducted over a wide frequency band provides more information about thetissue interiors which help us to better understand the biological tissues anatomy physiology and pathology Over past fewdecades a number of impedance based noninvasive tissue characterization techniques such as bioelectrical impedance analysis(BIA) electrical impedance spectroscopy (EIS) electrical impedance plethysmography (IPG) impedance cardiography (ICG) andelectrical impedance tomography (EIT) have been proposed and a lot of research works have been conducted on these methodsfor noninvasive tissue characterization and disease diagnosis In this paper BIA EIS IPG ICG and EIT techniques and theirapplications in different fields have been reviewed and technical perspective of these impedance methods has been presented Theworking principles applications merits and demerits of thesemethods has been discussed in detail along with their other technicalissues followed by present status and future trends

1 Introduction

A living object such as an animal or plant is developedwith cells and tissues arranged in three dimensional arraysFor example the human body is a biological subject whichis a very complex structure constructed by several livingtissues [1] composed of the three-dimensional arrangementof human cells The biological cells containing intracellularfluids (ICF) cellmembraneswith orwithout cell wall are sus-pended in the extracellular fluids (ECF) and show a frequencydependent behavior to an alternating electrical signal Underan alternating electrical excitation the biological cells and tis-sues produce a complex bioelectrical impedance or electricalbioimpedance [2ndash4] which depends on tissue compositionand frequency of the applied ac signal [2ndash4] Thereforethe frequency response of the electrical impedance of thebiological tissues is highly influenced by their physiologicaland physiochemical status and varies from subject to subjectEven the complex bioelectrical impedance varies from tissue

to tissue in a particular subject and also varies with the changein its health status [5 6] depending on the physiologicaland physiochemical changes occurred in the tissues healthHence the studies on complex bioimpedance of a tissue canprovide a lot of information about its anatomy and physi-ology Moreover as the bioelectrical impedance of a bodytissue depends on the signal frequency the multifrequencystudies on the electrical impedance of the biological tissuescan potentially be used for the noninvasive investigations oftheir physiological or pathological properties

Electrical impedance based noninvasive tissue char-acterizing techniques like bioelectrical impedance analy-sis (BIA) [7ndash21] electrical impedance spectroscopy (EIS)[22ndash32] electrical impedance plethysmography (IPG) [3334] impedance cardiography (ICG) [35ndash37] and electricalimpedance tomography (EIT) [38 39] are being used tostudy the frequency response of the electrical impedance ofbiological tissues But EIS is found more popular in severalfields of application compared to BIA IPG and ICG as it

Hindawi Publishing CorporationJournal of Medical EngineeringVolume 2014 Article ID 381251 28 pageshttpdxdoiorg1011552014381251

2 Journal of Medical Engineering

provides the impedance variations over frequencies AlsoEIS has been studied for the noninvasive characterizationof biological as well as nonbiological materials in frequencydomain whereas BIA IPG and ICG are used in biologicalfields only On the other hand the BIA IPG and ICG areapplied on biological tissues and generally at a particularfrequency BIA IPG and ICG all are impedance analyzingtechniques which provide the impedance values of the tissuesample as a lumped estimation whereas the bioelectrical EIScalculates and analyzes the electrical impedance at differentfrequencies which enable us to obtain not only the impedancevalues of the tissue sample as a lumped estimation at asuitable frequency (generally 50 kHz) but also it providesthe information to understand several complex bioelectricalphenomena like dielectric relaxation and dielectric disper-sions The information about the dielectric relaxation [40ndash42] and 120572 120573 and 120574 dispersions [40ndash42] enables us to analyzeand understand the complex bioelectrical phenomena [40ndash42] occurring in cells and tissues under an alternating electriccurrent signal BIA EIS IPG and ICG investigate the tissueproperties by assessing the lump impedance parametersobtained from the boundary voltage current measurementOn the other hand EIT provides a spatial distribution (2D or3D) of the impedance profile of a domain under test using aset of boundary voltage-current dataTherefore EIT is foundwith the potential of visualizing the tissue physiology andpathology in terms of tomographic images of the electricalimpedance distribution and hence it has been applied inseveral applications

The paper has discussed about the BIA EIS IPG ICGand EIT techniques and reviewed their technical perspectivealong with their application in different fields The workingprinciple applications merits and demerits of all thesemethods have been discussed in detail in this paper alongwith their other technical issues The paper discusses aboutthe present status challenges and future directions of theimpedance methods

2 Bioelectrical Impedance

21 Biological Tissues and Their Bioelectrical Impedance Theanimals or plants are the living subjects which are developedwith cells and tissues arranged in three-dimensional (3D)space Animals and plants are developed with animal tissuesand plant tissues respectively The animal tissues and theplant tissues are composed of the 3Darrays of animal cells andplant cells respectively Human body is a complex biologicalstructure composed of several tissues [1] composed of the3D arrays of cells All the biological cells such as animalcells and plant cells contain intracellular fluids (ICF) and cellenvelop (cell membrane for animal cells and cell membraneand cell wall for plant cells) and are suspended in extracellularfluids (ECF) The animal cells and plant cells suspended inextracellular fluids (ECF) show a different behavior to analternating electrical signal producing a complex electricalimpedance [2ndash4] which is called bioelectrical impedance orelectrical bioimpedance

The bioelectrical impedance depends on the tissue com-position as well as the frequency of the applied ac signal [2ndash4] and hence the frequency response of the bioelectricalimpedance of the biological tissues not only depends ontheir physiological and physiochemical status but also itvaries with applied signal frequency Moreover the bioelec-trical impedance varies from tissue to tissue and subject tosubject even if it varies for the measurement directions inanisotropic tissue structures Even the complex bioelectricalimpedance changes due to the variation in tissue structurecomposition and health status [5 6] Hence by studying theimpedance analysis of biological tissue one can obtain a lot ofinformation about the tissue anatomy and tissue physiologyTherefore the complex electrical impedance analysis of bio-logical tissues is found to be an efficient tool for noninvasiveinvestigations of physiological or pathological status

22 Origin of the Bioelectrical Impedance The biologicalsubjects are developed with biological tissues which aredeveloped with biological cells arranged in a very complexthree-dimensional arrays suspended in extracellular matrixcalled an extracellular fluids (ECF) [5] In animal tissuethe cells are composed of intracellular fluids (ICF) cellmembranes (CM) which are suspended in extracellular fluids(ECF) whereas the plant cells are composed of ICF CM andcell wall (CW) and are suspended in ECF

ICF CM and ECF all are composed of different materialshaving their distinguished electrical properties and thereforeeach of these cell and tissue components respond distin-guishably to an alternating current signal The intracellularfluids are composed of the cytoplasm and the nucleusThe cytoplasm and the nucleus are mostly made up ofsolution of proteins different chemicals salts and watersand hence these materials are electrically conducting Theextracellular fluids are alsomade up of electrically conductingmaterials As the intracellular fluids and extracellular fluidsin biological tissues are composed of ionic solution and thehighly conducting materials they provide highly conductingpaths (low resistive paths) to the applied current signal [34]But the cell membranes (Figure 1) of the cells in a biologi-cal tissue are composed of electrically nonconducting lipidbilayers (Figure 1(a)) sandwiched between two conductingprotein layers [35 36] and form the protein-lipid-protein(P-L-P) structures In the P-L-P structures (Figure 1(b)) thehydrophobic tails do not absorb the water and hydrophilicheads are attached to the protein layers The P-L-P [35 36]sandwiched structure (Figure 1(c)) provides a capacitance[34] to the applied alternating current signal and contributesto a capacitive reactance

As a result the overall response of the biological tissuesto an alternating electrical signal applied to it producesa complex bioelectrical impedance (119885119887) [2ndash4] which is afunction of tissue composition as well as the frequency ofthe applied ac signal [2ndash4] The impedance 119885119887 is a complexquantity and varies with tissue structure tissue compositiontissue health and signal frequency Therefore the impedancevaries from subject to subject tissue to tissue and sometimewith measurement direction within the same tissue Even

Journal of Medical Engineering 3

Hyd

roph

obic

(non

pola

r) ta

ils

heads

2D representation of P-L-P bilayer


ProteinHydrophilic zone

Hydrophilic zone

Hydrophobic zone

Hydrophilic (polar)

(a)

Hydrophobic(nonpolar) tails

Phospholipid units

Hydrophilic(polar) heads

(b)

Hydrophilic(polar)heads

Protein-lipid-proteinsandwiched structure

ProteinsPhospholipid units


(c)

Figure 1 Cell membrane structure of the biological cells (a) 2D and 3D model of P-L-P layers of a isolated cell membrane part (b) 2Dmodel of the lipid bilayers in cell membranes with hydrophilic (polar) heads and hydrophobic (nonpolar) tails and (c) 2D model of theprotein-lipid-protein sandwiched structure of a cell membrane

the bioimpedance of a tissuemay vary from other parts of thesame tissue due to the variation in tissue composition tissuephysiology and tissue pathology or tissue health

23 Bioelectrical Impedance of Diseased Tissue As the tissuephysiology and pathology change with the tissue health thebioelectrical impedance also varies from a healthy tissue toa diseased tissue Thus the bioelectrical impedance changeswith the variations in tissues health status [5 6] such asswelling disease and infection Similarly the impedance ofa normal tissue is found more than a tumor or canceroustissue as the cancerous tissue needs more blood to continueits rapid and uncontrolled growth which needs more bloodAs the blood is good conductor of electricity the canceroustissue containing more blood shows a less impedance path tothe electrical current

24 Frequency Response of Bioelectrical Impedance The fre-quency response of the bioelectrical impedance depends ontheir anatomical physiological and pathological status ofthe biological tissues Therefore the studies on the electricalbioimpedance of a tissue can provide a lot of informa-tion about its anatomy and physiology As the response of

the electrical bioimpedance varies with signal frequencymultifrequency impedance analysis provides more informa-tion about the tissue properties which help in the better tissuecharacterization Therefore the multifrequency impedanceanalysis of the biological tissues is found very promising forthe noninvasive investigation of physiological or pathologicalstatus As a result a lot of research studies have been con-ducted on the single frequency impedance analysis as well asthe multifrequency impedance analysis on biological tissuesto investigate their physiological or pathological health statusnoninvasively for tissue characterization as well as for thediagnosis of a number of diseases

3 Bioelectrical Impedance Analysis (BIA)

31 Impedance Measurement Electrical impedance of a par-ticular part of an object (volume conductor) is estimatedby measuring the voltage signal developed across that bodypart by injecting a constant current signal to the objectMathematically the impedance (119885) is measured by dividingthe voltage signal measured (119881) by the current signal applied(119868) 119885 is complex quantity and it will have a particularphase angle (120579) depending on the tissue properties Inelectrical impedance measurement process the bioelectrical


Voltmeter

Current source

Biological tissue sample

I

V

(a)

Current source

Voltmeter


I

V

(b)

Figure 2 Bioelectrical impedance measurement in technique (a) impedance measurement using two-electrode technique (b) impedancemeasurement using four-electrode technique

impedance of a body tissue is measured by injecting alow amplitude low frequency alternating current (generallysinusoidal) to the tissue through an array of surface electrodesattached to the tissue surface (tissue boundary)The alternat-ing current is applied to avoid the tissue damage and hencethe bioimpedance measurement is never conducted with thedirect current signal

In BIA the bioimpedance 119885ang120579 is found as the transferfunction of the SUT and thus the 119885ang120579 is calculated bydividing the voltage data (119881ang1205791)measured by applied current(119868ang1205792) as shown in

119885ang (120579) =

119881ang (1205791)

119868ang (1205792)

(1)

32 Two-Electrode and Four-Electrode Methods The bioim-pedance measurement process is conducted by either the twoelectrode or four-electrode methods In both the methodsthe surface electrodes through which the current signalis injected are known as the current electrodes or thedriving electrodes (as shown by the red colored electrodesin Figure 2) and the electrodes on which the frequencydependent ac potential (119881(119891)) is measured are called voltageelectrodes or sensing electrodes (as shown by blue coloredelectrodes in Figure 2)

As the name tells the two-electrodemethod (Figure 2(a))uses only two electrodes for impedance measurement andhence the current signal injection and voltage measurementare conducted with same electrodes The two-electrode-method therefore suffers from the contact impedanceproblem and the measurement data contains the voltagedrop due to the contact impedance In the four-electrodemethod (Figure 2(b)) two separate electrode pairs are usedfor current injection and voltage measurements and hencethe four-electrode method is found as an impedance mea-surement method with a linear array of four electrodesattached to the SUT as shown in Figure 2(b) The four-electrode method injects a constant amplitude current signalto the SUT through the outer electrodes called currentelectrodes or the driving electrodes (red colored electrodes inFigure 2(b)) and the frequency dependent developed voltagesignals are measured across two points within the currentelectrode through the inner electrodes called voltage elec-trodes or the sensing electrodes (blue colored electrodes inFigure 2(b))

33 Bioelectrical Impedance Analysis Bioelectrical impe-dance analysis (BIA) is a technique in which the body com-position of a biological object is analyzed by measuring itsbioelectrical impedance Hence the bioelectrical impedanceanalysis measures the bioimpedance of the tissue which isproduced inside the biological object when an alternatingcurrent tends to flow through it and hence it is found as afunction of tissue properties as well as the applied currentsignal frequency As explained earlier like other electricalimpedance measurement techniques in bioimpedance mea-surement methods a constant sinusoidal current is injectedto the subject under test (SUT) or a biological tissuesample and the voltage developed is measured using four-electrodemethod [5] or four-probemethod and the electricalimpedance of the SUT is calculated In four-electrodemethodof bioimpedance measurement process a constant amplitudealternating current is injected through two electrodes (outerelectrodes) called current electrodes and the voltages aremeasured on other two electrodes (inner electrodes) namedas voltage electrodes as shown in Figure 2 [5] BIA determinesthe bioelectrical impedance of a particular body part attachedwith surface electrodes [5] Bioelectrical impedance of asubject can be used to calculate an estimate of the bodycomposition

A bioelectrical impedance analysis (BIA) [7ndash14] is a tech-nique in which the body composition [14ndash21] of a biologicalobject is analyzed by measuring its electrical impedancecalled bioelectrical impedance or electrical bioimpedanceDrWilliamMills MD an Admiral in the USNavy initiatedthe study onBIA inMountMcKinley Alaska in 1981 to assessthe hydration status of soldiers in high altitude cold weatherenvironments The paper published by Hoffer et al [7] in1969 indicated a procedure to predict the total body waterby measuring the hand to foot whole body BIA and with theencouragement of Jan Nyboer the Mount McKinley soldierhydration project was started Shortly thereafter Lukaski et al[8] at the USDA in Grand Forks ND published the first to apaper on BIA and body composition

34 Body Compositions Human body is developed withthe complex arrangement of different body tissues consistsof water protein fat and minerals called the body com-positions The human body is composed of 64 water20 protein 10 fat 1 carbohydrate and 5 mineralsapproximately [15] as shown in Figure 3(a) Over 90 of


Water 64

Protein20

Fat10

Minerals5

Carbohydrate 1

Human body compositions

(anatomy model)

(a)

Oxygen 65

Carbon 18

Hydrogen10

Others7


(chemical model)

(b)

Figure 3 Compositions of human body (a) anatomy model of the human body composition percentage of fat protein minerals and liquidin an adult human subject (b) chemical model of the human body composition percentage of different elements in the body composition ofhuman subjects

the mass of the human body is made by the just threeelements namely oxygen (65) carbon (18) and hydrogen(10) among 118 elements discovered so far (Figure 3(b))The human body compositions are largely divided into twogroups body fat and lean body The lean body is divided toprotein mineral and body water Protein is a main elementof muscles whereas the minerals are found mostly in bones[43] Total body water (TBW) consists of intracellular water(ICW) and extracellular water (ECW) ICW are found withinthe cells and tissues and give cell volume and tissue volumeswhereas the ECW is composed of blood lymph and so forth[43] Body cell mass is the sum of ICW and metabolicallyactive tissues [43] The body-composing constituents andtheir explanation and function are presented in Table 1 Bodycomposition (Figure 3(a)) of human subject is essentiallyto be estimated to study and diagnose the normal healthdiseases and the other abnormal health status Human bodymass is developed with several elements (Figure 3(b)) inform of water cells and tissues Human body mass consistsof tissues with high conductivity called lean body mass(LBM) and tissues with low conductivity called body fat (BF)Generally an unhealthy body composition refers to carryingtoo much fat in comparison to lean tissue (muscle) andthe more the bodyrsquos fat the more the health risks In factunhealthy body composition often leads to obesity creatingmany other health complications like heart disease strokehigh blood pressure high cholesterol type 2 diabetes backpain and so forth

Table 1 Body-composing constituents and their explanation andfunction [43]

Body-composingconstituents

Explanation and function

Intracellular water(ICW)

Body water which exists inside of cellmembrane

Extracellular water(ECW)

Body water that exists outside of cellmembrane (blood interstitial fluid etc)

Body waterThe sum of intracellular and extracellularwater

ProteinMain element which composes soft leanmass together with water

Soft lean mass (SLM)Skeletal muscle and smooth muscleMaintaining body function

Minerals Composing bones and electrolytes

Lean body mass (LBM)

Composed with soft lean mass andmineralsThe amount of body weight minus bodyfat mass

Body fatThe amount of body weight minus leanbody mass

Weight

The sum of lean body mass and body fatmass Standard weight (kg)adult male height (m) times height (m) times 22female height (m) times height (m) times 22


Men Women

Water 55ndash65 Water 55ndash65

Fat 15ndash20

Minerals 58ndash60

Protein 16ndash18

Fat 20ndash30

Minerals 55ndash60

Protein 14ndash16

Figure 4 Optimal ratio of body composition

(a)

(ICW) asymp 44

(ECW) asymp 29

Bone minerals asymp 7

FM = weight minus FFM

Visceral proteins

Intracellular water

Extracellular water

Fat free mass (FFM) Total body

water (TBW)

Body cell mass (BCM)

(b)

Figure 5 Human body as a conducting cylinder and its body composition (a) human body assumed as a conducting cylinder in BIA (b)body composition schematic diagram of fat-free mass (FFM) total body water (TBW) intracellular water (ICW) extracellular water (ECW)and body cell mass (BCM) [9]

35 Optimal Ratio of Body Composition for Male and FemaleThe optimal ratio of body composition is shown in Figure 4In human body the following relations are held for bodycompositions

(1) lean body massprop soft lean massprop body water

(2) lean body massprop 1percent of body fat

(3) lean body mass varies directly as soft lean mass orbody water but inversely as body fat mass

(4) because of weight = lean body mass + body fat

36 BIA and Body Composition Calculation

361 Fat-Free Mass (FFM) or Lean Body Mass (LBM) Thefat free mass (FFM) or the lean body mass (LBM) or musclemass or body cell mass (BCM) [9] is developed with muscleas well as the metabolically active tissue in your organs [9]Fat-free mass is comprised of the nonfat components ofthe human body Skeletal muscle bone and water are allexamples of fat-free mass Several equations are proposedto calculate the fat-free mass The schematic diagram of thebody composition such as fat-free mass (FFM) total bodywater (TBW) intracellular water (ICW) extracellular water(ECW) and body cell mass (BCM) is shown in the Figure 5


Current conduction through the muscle tissuesImpedance of the muscle tissue is very low

Electricity conduction is high

Muscle tissues with more water

(a)

Current conduction through the fat tissuesImpedance of the fat tissue is very high

Electricity conduction is less

Adiposetissues with

less water

(b)

Figure 6 Current conduction through body tissues (a) through muscle tissue and (b) through fat tissue

The following equations have low standard errors forpredicting FFM and is appropriate for the general population[44]

for females

Fat-Free Mass (kg)

= 0475 times [

(ht (cm))2

119877 (ohms)] + 0295 times wt (kg) + 549

(2)

for males

Fat-Free Mass (kg)

= 0485 times [

(ht (cm))2

119877 (ohms)] + 0338 times wt (kg) + 352

(3)

362 Body Fat (BF) The body fat is the amount of fat tissuewithin a body that is the total adipose tissues without anymuscle tissue or body fluids or electrolyte [9] It is the totalbody mass minus the FFM

Body Fat (BF) (kg) = Total Body Mass (TBM) (kg)

minus Fat-Free Mass (kg) (4)

363 Total BodyWater Body (TBW) Several formula of totalbody water body (TBW) has been proposed by differentresearch groups Body water consists of intracellular water(ICW) and extracellular water (ECW) [18] Total body waterbody (TBW) is given by [18]

TBW (kg) = 184 + 045(

height2 squareresistance

)

+ 011 (weight)

(5a)

According to Lukaski et al [8] and the JawonMedical Inc[43] South Korea the formula is given by

TBW (kg) = 0377(

height2

resistance) + 014Weight

minus 008Age + 29Gender + 465

(5b)

TBW (kg) = A times (

height2

resistance) + B timesWeight

minus C times Age + D times Gender

(5c)

where A B C D and E are the constants

37 Body Tissue Conductivity Fat tissues are the adiposetissues consisting of fat cells which are a unique type ofcell containing electrically low conductive materials Thefat tissue impedance is high as the conductivity of the fattissue is low (Figure 6(a)) On the contrary lean tissuescontain intracellular and extracellular fluid and electrolytesTherefore as the conductivity of the lean tissue is high theimpedance is low in lean tissues (Figure 6(b)) Impedanceis thus proportional to TBW Under an alternating currentinject to the body the current signal flows along withpaths contain more water within the body since it has highconductivity Depending on thewater content in the body theimpedance value changes for body water body fat and bodymuscle The ratio of these two types of tissues is reflected onthe electrical characteristic and the impedance value [45]

38 BIA for Body Composition and Health The Necessity Ina healthy human body a certain ratio of the compositionsis maintained to sustain a good health Hence the bodycomposition assessment is essential to study the humanhealth as well as to diagnose a number of diseases like obesityedema and protein-deficient malnutrition as well as to assessthemetabolic status in human bodyThough there are severalmethods to assess the body compositions like skin-fold test


BIA

(a)

Right hand Right foot

One of the driving electrodes is placed on the first joint of the middle finger of the right hand

One of the driving electrodes is placed on the base of the second toe of the

right foot

One of the sensing electrodes is placed on

an imaginary line bisecting the ulnar head

(bone on little finger side of wrist of the right

hand)

One of the sensing electrodes is placed on an imaginary line bisecting the medial mallealus (bone on big toe side of ankle of the right foot)

(b)

Figure 7 Body composition assessment using BIA (a) BIA current injection and voltage measurement schematic (b) electrode connectionin impedance measurement procedure of the BIA

[21 45] underwater weighing [21 45] waist-to-hip ratiomethod [46] and dual energy X-ray absorptiometry (DEXA)[47] still the accurate noninvasive low cost fast and easymethods are always desired in medical and clinical fields forbody composition assessments In this direction bioelectricalimpedance analysis (BIA) which is able to properly identify asubjectrsquos health risk of excessively high or low body fatnessis being studied and researched by a number of researchgroups including doctors clinicians medical practitionersand biomedical engineers It is a growing and promisingtechnique that ranks similar to skin-fold measurement in itsaccuracy precision and objectivity [48] BIA is used to studyanalyze and evaluate the body fat and body fat freemass BIAalso shows many other health parameters such as your BMI(body mass index) and total body water (TBW)

39 Advantages of BIA Compared to the other noninvasiveprocedures to assess the body compositions the BIA tech-nique has the following advantages

(i) noninvasive safe fast low cost portable easy toconduct hazards free and safe technique

(ii) measures fat-free mass and calculates fat mass(iii) calculates of body cell mass total body water intra-

cellular water and extracellular water(iv) BIA devices are light in weight portable and can be

used at the bedside(v) BIA devices can be used to assess total body water in

individuals with altered metabolic function(vi) excellent consistency for repeated measurements

310 BIA Procedures BIA is studied with an instrumentcalled body composition analyzer (BCA) (Figure 7(a)) whichis dedicated to inject a constant current and measure theelectrical impedance of the body (119885119861) using standard four-electrode method (Figure 7(b)) As explained earlier in four-electrode method based BIA procedure a low and constantamplitude (le1mA) alternating current (generally 50 kHz)

is injected through two electrodes (outer electrodes) calledcurrent electrodes and the voltages aremeasured onother twoelectrodes (inner electrodes) named as voltage electrodes asshown in Figure 7(b) The resistance (R) or resistivity (120588) iscalculated from the impedance (119885119861) and using the resistivityor conductivity various body compositions are calculated asshown in Section 36

311 Assumptions in BIA Calculations The BIA assumesthe human body as a cylindrical homogenous conductor(Figure 8)whose impedance is proportional to the length andinversely proportional to the cross-sectional area of the base(Figure 8)

Though in practical case the human body is differentfrom the assumptions mentioned above but for ease ofcalculation the BIA formulation procedures generally havethe following assumptions

(1) human body is a cylinder determined by height andweight

(2) body composition is homogenous and evenly dis-tributed

(3) there are no individual differences and variationswithin the body compositions

(4) there are no changes in the environment (tempera-ture) body heat and stress

The BIA procedure applies an alternating current whichexperiences a bioelectrical impedance exerted by the cellsand extracellular fluids As the cell membranes are capacitivein nature the capacitive reactance produced by the electriccurrent applied selectively allows the current pass throughit depending on the signal frequency and hence the currentpaths (Figure 8) The low frequency current passes throughthe extracellular fluids as the cell membrane reactancedoes not allow the low frequency current to pass throughit whereas the high frequency current penetrates the cellmembranes and passes through both the extracellular fluidsand the cells (membranes and intracellular fluids) Thus by


AZ =

120588L

A

Volume = 120588height2

Z

or Z =120588L2

volumeL

Sinusoidal current Cells

Low frequency current

Highfrequency

current

Extracellularfluids

Figure 8 Impedance of the human body assuming it as a homoge-neous cylindrical volume conductor

applying the alternating current at a particular frequencyBIA procedure can assess the amount of extracellular water(ECW) intracellular water (ICW) and total body water(TBW = ECW + ICW)

312 BIA Outcomes The BIA results reveal several informa-tion about the body composition and the body health suchas

(i) LBM and BF ratio how much lean body mass (LBM)versus body fat (BF) a person has

(ii) Toxic burden how much the toxic burden of thehuman subject is

(iii) Water intake whether the person is drinking ade-quate amount of water or not

(iv) Burning calories how many calories the person isburning at rest

(v) Loosing and gaining of LBM or BF whether thesubject is losing or gaining muscle mass or fat

(vi) Health status how healthy the personrsquos cells are

313 BIA Preparations and Precautions The BIA procedureis conducted on the patients by attaching few electrodeson their body and the entire experimentation is completedwithin 3-4 minutes The BIA experimentation is conductedon the patients either in lying or standing condition withthe electrodes attached to the body and a small amountof alternation current is injected and the potentials aremeasured The amplitude of the electrical current is so smallthat it cannot be felt at all The BIA procedures are conductedon the subject who should not

(1) eat or drink caffeine for 4 hours prior to yourappointment

(2) exercise for 12 hours prior to the personrsquos appoint-ment

(3) wear tight clothing or pantyhose(4) apply lotion on the personrsquos hands and feet prior to

your appointment and(5) let the person take his diuretic medication within 6

hours prior to your appointment

The BIA procedures are conducted on the subject whoshould

(1) remove all metal jewellery prior to doing the assess-ment and

(2) drink 2ndash4 glasses of water within 2 hours of yourappointment

314 BIA Procedure Cost Bioelectrical impedance analysis(BIA) is a simple safe noninvasive and painless procedurewhich is conducted by a single person a doctor medicalpractitioner clinician nurse or even a laboratory expertTheapproximate cost of a BIA experiment is around $35ndash$65 [49]TheBIA test fee inHealthyDirectionsNutritionTherapy andCounseling Center of Doylestown Hospital USA is $25 [50]

315 Single Frequency BIA and Dual Frequency BIAResearches show that the alternating current BIA canbe performed in either single frequency current (calledsingle frequency BIA or SFBIA) or dual frequency current(called dual frequency BIA or DFBIA) [51] GenerallySFBIA is performed with 50 kHz and DFBIA is performedwith 50 kHz and 200 kHz TBW can be measured usingSFBIA and TBW and ECW can be calculated by DFBIA Bymeasuring the impedance (119885119861) at 50 kHz and 200 kHz andby applying predictive equations (5a) (5b) (5c) and (6)both extracellular water (ECW) and TBW respectively arecalculated and the intracellular water (ICW) is deduced [45]The extra-cellular mass (ECM) and body cell mass (BCM)are found from ECW and ICW respectively [45] Severalequations are proposed and used to predict fat-free mass(FFM) ECW can be found related to extracellular mass(ECM) and ICW to body cell mass (BCM) The equation forSFBIA and DFBIA is shown in (5a) (5b) (5c) and (6)

TBW (Liters) = [

(03963 times height2)impedance (50 kHz)

]

+ (0143 times weight) + 84

(6)

ECW (Liters) = [


]

+ (006895 times weight) + 3794

(7a)

TBW (Liters) = [


]

+ (018782 times weight) + 8197

(7b)

4 Electrical Impedance Spectroscopy (EIS)

Electrical impedance spectroscopy (EIS) [52ndash54] estimatesthe complex electrical impedance (119885(119891)) and its phaseangle (120579(119891)) of a subject under test (SUT) at differentfrequency points 119891119894 (1198911198941198911 1198912 1198913 119891119899) by measuring thesurface potentials (119881(119891)) developed for a constant current


injection (119868(119891)) at the boundary through a linear array of thesurface electrodes attached to the SUT (Figure 9) In EIS afrequency dependent constant amplitude sinusoidal current(119868(119891)) is injected through either two-electrode method orfour-electrode method As explained earlier in 2-electrodebased EIS and 4-electrode based EIS methods the surfaceelectrodes through which the current signal is injected areknown as the current electrodes or the driving electrodes (asshown by the red colored electrodes in Figure 9) and the elec-trodes onwhich the frequency dependent ac potential (119881(119891119894))ismeasured are called voltage electrodes or sensing electrodes(as shown by blue colored electrodes in Figure 9) Thereforein EIS the frequency dependent electrical bioimpedance119885(119891119894) is found as the transfer function of the SUT and thusthe 119885(119891119894) is calculated by dividing the voltage data (119881(119891119894))measurement by applied current (119868(119891119894)) as shown in

119885 (119891119894) =

119881 (119891119894)

119868 (119891119894)

(8)

EIS can be potentially used as a nondestructive eval-uation technique [55] for a number of applications in thefield of science engineering and technology The EIS-basedfrequency response studies on the electrical impedance ofany material can provide its structural and compositionalproperties as well the frequency response of the materialproperties which can be potentially used for nondestructivematerial characterization [52ndash54]Therefore the EIS has beensuitably applied in several fields such as electrochemistryand chemical engineering [54 56ndash61] material engineering[62ndash71] biomedical engineering [72ndash76] civil engineering[77ndash79] wood science [80 81] plant physiology [82ndash84]microfluidics [85 86] material engineering [87ndash90] fuel celltechnology [91] andMEMS and thin films [92 93] and so on

5 Impedance Plethysmography (IPG)

Impedance plethysmography (IPG) is a electrical impedancebased noninvasive medical diagnostic procedure which mea-sures small changes in the blood volume in terms of itselectrical bioimpedance of a body part say chest calf or otherregions of the body to study the tissue health condition ofthe patient The impedance measured in IPG provides theinformation about the tissue health of the body tissue andhence it can be suitably used to indirectly study and analyzethe tissue health and function of a patient under test There-fore the IPG can be considered as a bioimpedance analyzingtechnique that measures the change in blood volume for aspecific body segment bymeasuring the electrical impedancewhich changes with a change in the blood volume

51 IPG Procedure In IPG four conductive bands are tapedaround the body part or a limb and the limb impedancechange due to the blood circulation is measured to calculatethe blood volume changes (Figure 10(a)) A low amplitudelow frequency (50 kHz) ac current is passed through twoouter electrodes and the change in electrical impedance ismeasured (Figure 10(b)) across the inner electrodes In early

Subject under test (SUT)

Voltage electrodesCurrent

electrodeCurrent electrode

Stop

Impedance analyzer model ZXR00

Start

Figure 9 Electrical impedance spectroscopy (EIS) studies on bio-logical materials with four-electrode technique using an impedanceanalyzer

1950rsquos Nyboer [94 95] proposed the impedance plethys-mography equation to evaluate the volumetric change in thebody parts in terms of impedance But the formula proposedby Swanson and Webster 1976 [96] is more simpler bothmathematically and computationally The formula works onthree assumptions the expansion in the artery is uniformblood resistivity (120588119887) does not change and current fluxes areparallel to the artery (Figure 10(b)) In the theory of IPG foreach pressure pulse produced due to the blood flow throughlimb having an artery with an volume 119881 and cross sectionalarea119860 an extra amount blood with a volume of Δ119881 enters tothat limb by increasing the limb volume from 119881 to 119881 + Δ119881

and consequently a shunting impedance (119885119887) is assumed tobe produced Therefore the (119885119887) is given by [96 97]

119885119887 = 120588119887

119871

Δ119860

(9)

Hence the artery volume change is given by [96 97]

Δ119881 = 119871 sdot Δ119860 = 120588119887

1198712

119885119887

(10)

For each pressure pulse the area of the artery (119860)

increases from 119860 to 119860 + Δ119860 (Figure 10) Consequently theartery impedance (119885119887) is reduced by adding an impedance(Δ119885) produced by the Δ119860 and this 119885119887 will be connected inparallel to 119885 [96 97]

As we measure the Δ119885 rather that 119885119887 the 119885119887 is requiredto be replaced by Δ119885 Now the blood impedance change Δ119885is given by [96 97]

Δ119885 = (119885119887 119885) minus 119885 = minus

1198852

119885 + 119885119887

(11)


Current electrodes

Body part

Current source

Voltmeter

Voltageelectrodes

I

V

(a)

Body part

Zb

Z

Z

A

ΔZ

ΔA

L

(b)

Figure 10 The schematic of the impedance measurement in IPG with four-electrode method and the calculation of limb volume in terms ofbioimpedance (a) impedance measurement of body part with four-electrode method (b) volume change in body part due to blood pressurepulse and corresponding impedance variation due to the volume change

Now as 119885119887 ≫ 119885 the 119885119887 is given by [96 97]

1

119885119887

cong minus

Δ119885

1198852 (12)

Therefore the increases in the limb volume ΔV is givenby [96 97]

Δ119881 = minus120588119887

1198712Δ119885

1198852 (13)

If all the assumptions are valid then the volumetricchange in a body part due to blood flow can be calculatedfrom the impedance measurement using IPG technique

Peterson [98] has conducted the experiments conductedon an impedance plethysmograph suitable for ambulatoryblood pressure monitoringThe impedance plethysmographdeveloped and used in the experiment worked as an ambula-tory blood pressuremonitor (ABPM) contained four circularcopper band electrodes attached to the upper left forearm of avolunteer subject In ABPM the impedance wasmeasured byusing a signal generator connected to the outer two bands forelectric current injection and an amplifier circuit connectedacross the inner two electrodes for voltagemeasurement [98]

52 Advantages Compared to the other techniques theImpedance Plethysmography (IPG) has a number of advan-tages as listed below

(i) noninvasive(ii) low cost(iii) fast process(iv) portable and easy to operate and dry type testing

procedure(v) bedside measurement and ambulatory measurement

are possible(vi) measurement in ICU is possible(vii) current density in IPG with ring electrodes is more

uniform compared to other four electrode methodsusing spot electrodes

(viii) skin electrode impedance effect is less and skinimpedance can be further reduced by applying theelectrode gel

(ix) electrical signal in IPG is easily to control process andacquire

(x) IPG is less temperature dependent(xi) surrounding humidity effect in IPG is less(xii) applicable for healthy persons and patients of all age

groups(xiii) IPG helps doctors and clinicians to measure changes

in venous blood volume as well as the arteries pulsa-tions

(xiv) compared to venography which is invasive andrequires a skilled person to perform and interpretaccurately IPG is an easy to perform and understand

(xv) ICG helps a doctor to detect deep vein thrombosis(DVT)

(xvi) IPG is found 96 sensitive and 98 specific in thediagnosis of arterial occlusive disease and more than85 sensitive in the diagnosis of DVT and valvulardiseases

53 Limitations IPG has the following limitations

(i) electric current is injected to the subject(ii) electro skin impedance error is introduced

54 Applications Due to the number of advantages of theIPG it has been used for several application as follows

(i) blood volume measurement(ii) detection of deep vein thrombosis (DVT)(iii) detection of venous and arterial insufficiency(iv) screening of the patients who are likely to have blood

clots in legs(v) detection of the source of blood clots in the lungs

(pulmonary emboli)


Outer electrodes

injects current signal

Inner electrodes measures the

potentials

Xiphoid

Sternum

Xiphisternal joint

(a)

8-electrode connection of ICG

copy B Bo SramekSraSSrSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSS mekkkkkkkkkkkkkkkkkkkkkkkkkkkkkkkkkkkkkkkkkekkekkee

Vena cavae

HF measurementcurrent

Thoracic aorta

TEB and ECGsensing

(b)

Figure 11 The schematic of the impedance measurement in ICG with four-electrode method (a) impedance measurement withband electrode (photograph courtesy Ref [100]) and (b) impedance measurement with band electrode (photograph courtesyhttpwwwhemosapienscomtebhtml (with permission of B Bo Sramek) Retrieved on 12122013)

(vi) indirect assessment of central and peripheral bloodflow

(vii) detect blood flow disorders such as venous andarterial occlusive diseases (estimate severity)

(viii) early stage arteriosclerosis(ix) functional blood flow disturbances migranes(x) general arterial blood flow disturbances

6 Impedance Cardiography (ICG)

61 Introduction Impedance cardiography (ICG) alsoreferred to as transthoracic electrical impedance plethys-mography is a technology which calculates the changes inblood volume in transthoracic region over time in termsof the changes in transthoracic impedance (Figure 11)called thoracic electrical bioimpedance (TEB) or 119885119900 Bodyimpedance is changed in human body due to the bloodcirculation caused by each heart rhythms and hence thetransthoracic impedance (119885119900) is influenced by its biologicalcomposition breathings and by the blood circulationand blood volume in the transthoracic region Thereforeanalyzing the119885119900 the health of the heart along with a numberof hemodynamic parameters can be evaluated ICG has beenresearched since the 1940s NASA developed the technologyin the 1960s [99]

62 ICG Procedure The ICG procedure is basically similarto the IPG procedure in which the impedance measure-ment is conducted by injecting a low amplitude constantalternating electric current (frequency range of 20 kHzndash100 kHz) [58] into the volume conductor and measuring thecorresponding voltage (Figure 11) The frequency dependentimpedance 119885(119891) is measured from the ratio of voltage 119881(119891)to current 119868(119891) applied usually the DC value is eliminatedand only the impedance variation Δ119885 is further examined[100]The current injection and the voltage measurement areconducted with four-electrode method either using the ring

electrode or the spot electrodes (common ECG electrodes)In ring electrode based ICG system four ring electrodes arerequired whereas the ICG with spot electrode configurationneeds eight ECG electrodes Hence all the ICG systems areoperated with four-electrode measurement technique usingeither four-band electrodes (Figure 11(a)) or 8 spot electrodes(Figure 11(b)) The 4-electrode and 8-electrode connectionsof ICG are shown in Figures 11(a) and 11(b) respectivelyIn ring electrode connection one electrode of the drivingelectrodes (the outer pair) is placed around the abdomenand the other is placed around the upper part of the neck(Figure 11(a)) For the sensing electrode pairs (inner electrodepair) one electrode is placed around the thorax at thexiphisternal joint and the other around the lower part ofthe neck (Figure 11(a)) In 8-electrode connections of ICGalso the current injection and voltage measurement are con-ducted using four-electrode method as the current is injectedthrough the two upper most electrodes and two lower mostelectrodes and the voltage data are collected across theinner two sets (each set contains two spot electrodes) ofthe electrode In recent times the band electrodes are oftenreplaced with spot electrodes such as normal ECG electrodes[100] A low magnitude low frequency ac signal is injectedby current (red in the figure) electrodes Voltage developedacross voltage electrodes due to the current injection throughthe current electrodes is measured and the average transtho-racic impedance (119885119900) across the transthoracic region andthe small change in impedance (Δ119885) due to blood flow arecalculated and monitored against time After getting the 119885119900119889119885119889119905 Δ119885 are calculated Analyzing the 119885119900 119889119885119889119905 Δ119885 thestroke volume cardiac volume and several hemodynamicparameters are calculated for noninvasive diagnosis of theheart and circulatory system

Figure 12 shows a typical thorax impedance curve (119885)its first time derivative (119889119885119889119905) the simultaneous electrocar-diogram (ECG) and phonocardiogram (PCG) curves [100]It is important to note that a decrease in impedance resultsin an increase in the 119910-axis magnitude and hence this signconvention describes the changing transthoracic admittance


R minus Z

Z

10Ωs

Calibration

ECG

PCG

1 2 3 4

Time (s)0

A B XYOZ

dZ

dt

dZ

dtmin

Figure 12 The transthoracic impedance curve [100] over time

Thus a decrease in impedance can be explained by an extraamount of blood (low impedancematerial) flow in the thoraxIt is also important to note that the polarity of the firstderivative curve (119889119885119889119905) is consistentwith the impedance (119885)curve

63 ICG andHemodynamic Parameters In ICG a number ofhemodynamic parameters are derived to study the hemody-namic health and functions including the following

(i) stoke volume(ii) cardiac volume(iii) stroke volumestroke volume index (SVSVI)(iv) cardiac outputcardiac index (COCI)(v) systemic vascular resistanceindex (SVRSVRI)(vi) velocity index (VI)(vii) thoracic fluid content (TFC)(viii) systolic time ratio (STR)(ix) left ventricular ejection time (LVET)(x) preejection period (PEP)(xi) left cardiac workindex (LCWLCWI)(xii) heart rate (HR)

631 Hemodynamic Parameter Calculation The followingsection represents the detail explanations and the formula ofthe hemodynamic parameters [100 101]Stroke Volume Stroke volume is very important in cardio-vascular physiology to determine the cardiac function andcardiac health and cardiac parameters such as cardiac output(CO) ejection fraction (CEF) and so forth Stroke volume ismeasured in milliliters per beat (mLbeat) which is almostequal for each ventricles (approximately 70mLbeat for anormal adult subject) SV is indexed to a patientrsquos body size bydividing by body surface area (BSA) to yield stroke index (SI)

Using the information acquired from the echocardiographyfor a given ventricle the SV is estimated by subtractingvolume of the blood in the ventricle at the end of a beat (calledend-systolic volume or ESV) from the volume of blood justprior to the beat (end-diastolic volume or EDV) Using theimpedance method the SV is calculated by the formula asdescribed below

(i) stroke volume = [120588]times [119871119885119900]2times Δ119885 (Nyboer Formu-

lae)(ii) stroke volume = [120588] times [119871119885119900]

2times 119889119885119889119905 times LVET

(Kubicek Formulae)(iii) stroke volume = [119881119885119900] times 119889119885119889119905 times LVET (Sramek

Formulae)

where 120588 is specific resistivity of blood which is a constant fora particular subject 119871 is transthoracic length 119885119900 is averagetransthoracic impedance LVET is left ventricular ejectiontime 119881 is volume of the transthoracic tissues involvedelectrically during the test = 1198713425Cardiac Volume The cardiac volume refers to the volume ofthe heart which is usually relating to the volume of bloodcontained within it at various periods of the hear cycleCardiac Output (CO) Cardiac output (CO) is the totalamount of blood ejected or pumped out by the left ventricle ofthe heart into the systemic circulation in one minute Hencethe CO is found as equal to the stroke volume times the heartrate Using the cardiac output of a human subjects one morehemodynamic parameter called the cardiac index (CI) iscalculated by dividing the COby the body surface area (BSA)The CO is measured in (Lmin or mLmin) and as the CI ismeasured in litres per minute per square metre (Lminm2)The CO is generally found as approximately 56 Lmin fora human male subject and 49 Lmin for a human femalesubject Thus the CO and the CI are found as

CO = (stroke volume times heart rate)CI = cardiac outputbody surface area

64 Advantages The impedance cardiography (ICG) has anumber of following advantages

(i) noninvasive low cost fast portable safe(ii) current distribution is better and less noise effect(iii) electrode-skin impedance effect being less

7 Electrical Impedance Tomography (EIT)

71 Introduction to EIT Electrical impedance tomography(EIT) [102ndash114] a computed tomographic image reconstruc-tion technique is a nonlinear inverse problem in whichthe electrical conductivity or resistivity of a conductingdomain (Ω) is reconstructed from the surface potentialsdeveloped by a constant current signal injected (Figure 13)at the domain boundary (120597Ω) A low frequency constantamplitude sinusoidal current is injected to the boundary (120597Ω)of the object domain (Ω) to be imaged within a volume


120597Ω

Ω

VV

V

VV

I

Figure 13 A closed domain of interest under EIT scanning with aconstant current injection through driving electrodes and boundarypotential measurement on sensing electrodes

conductor surrounded by an array of surface electrodes andthe boundary potentials are measured using an electronicinstrumentation [115ndash123] Constant current signal is injectedto the closed domain under test (DUT) through the differentpairs of electrodes called driving electrodes and the voltagedata are collected from the other electrodes or electrode pairscalled sensing electrodes and the set of voltage current dataobtained from this noninvasive boundary measurements isused to reconstruct the conductivity distribution of the DUTBoundary data are then sent to the PC and the data areprocessed in PC to reconstruct the spatial distribution of theelectrical conductivity of the DUT using a computer programcalled image reconstruction algorithm [124ndash133]

EIT is a low cost portable fast noninvasive nonionizingand radiation free technique and hence it is found advanta-geous in several fields of applications application comparedto the other computed tomographic methods like X-ray CT[134ndash136] X-raymammography [137]MRI [138 139] SPECT[140 141] PET [142 143] ultrasound [144 145] and so forthEIT has been applied in a number of research areas suchas medical imaging clinical diagnosis [146ndash152] chemicalengineering [153] industrial process application [154 155]material engineering [156] microbiology and biotechnology[157 158] nondestructive testing (NDT) in manufacturingtechnology [159] civil engineering [160] earth science andgeophysics and geoscience [161] defense fields [162] archae-ology [163] oceanography [164] environmental engineering[165] and other fields of applied science engineering andtechnologies [166]

72 Physics of EIT In EIT when a constant current isinjected into DUT the current signal conducted throughthe domain produces current fluxes which induce potentialswithin the DUT The potential profiles developed by thecurrent conduction depend on the profiles of the currentfluxes which are influenced by the impedance profile ofthe DUT As the profile of the current conduction in

a homogeneous domain (Figure 14(a)) differs from thecurrent flux lines produced by the inhomogeneous DUT(domain with inhomogeneity) (Figure 14(b)) the voltageprofile of the homogenous DUT will be different fromthe domain with inhomogeneity Similarly the boundarypotential profiles will depend on the domain impedancedistribution and hence the information of the impedancedistribution of the DUT is hidden inside the boundarypotential data

73 Comparison with CT EIT is a computed tomographicimaging modality which uses an electrical energy (eitherby injecting electrical current or voltage signals) and thedeveloped potentials are collected at the domain boundarywhereas in X-ray CT X-ray beams are passed through theSUT at different angles called the projection angles and theattenuated X-ray beams are collected by the X-ray detec-tors The image reconstruction algorithm uses the measureddata in PC and reconstructs the spatial distribution of theelectrical impedance and the X-ray attenuation coefficientof the SUT in EIT and CT respectively Therefore the CTis a tomographic technique imaging with X-rays and EITreconstructs the tomographic images with electric current

74 Advantages of EIT EIT has several advantages over othercomputed tomographic imaging techniques used for medicalimaging applications as summarized below

(i) noninvasive(ii) radiation free(iii) nonionizing method(iv) fast data acquisition(v) high temporal resolution(vi) medically safe process(vii) low cost device(viii) portable device and easy to use(ix) suitable for bedside measurement and ICU monitor-

ing(x) suitable for ambulatory monitoring(xi) negligible patient preparation is required(xii) no precautions are required(xiii) no postexperimental discomfort(xiv) no postexperimental precautions and restrictions(xv) suitable for the patients of any age groups as well as

critically ill patients

75 Capacitance Tomography and Resistance Tomography InEIT the spatial distribution of the electrical conductivityis reconstructed from the boundary data collected by analternating current injection at the domain boundary Insome applications the electrical permittivity of the DUTis reconstructed from the voltage current data collected atthe boundary and the imaging modality is called electricalcapacitance tomography (ECT) [125 133 167 168] which is


I

V

V

V

V

V

V

V

V

V

VVV

V

1

2

3456

7

8

9

10

1112 13 14

15

16

(a)

I

V

V

V

V

V

V

V

V

V

VVV

V

1

2

3456

7

8

9

10

1112 13 14

15

16

(b)

Figure 14 Variation in boundary data profiles for a homogeneous domain and a inhomogeneous domain (a domain with inhomogeneity)due to the variation in the profiles of impedance distributions (a) domain without inhomogeneity (b) domain with inhomogeneity

generally used in industrial process application and mechan-ical and material engineering The electrical resistivity ofthe domain under test is sometimes reconstructed from theboundary data generated by a direct current injection inindustrial process imaging civil structure imaging subsur-face imaging and other applications of geotechnology Thismethod is known as the electrical resistance tomography(ERT) [169ndash173]

76 Difference EIT and Static EIT EIT image reconstructionsare conducted from the boundary data either with thedifference or dynamic imaging modality called the differ-ence imaging or another modality called static imaging Indifference imaging the impedance distribution of the DUTis reconstructed by comparing the data collected from thedomain with inhomogeneity and the data collected from thehomogeneous domain (reference data) whereas the staticimaging reconstructs the conductivity distribution from thedata collected from the inhomogeneous medium withoutusing the reference data setThe present EIT technologies aregenerally found applied in difference imaging modalities asthe static imaging seems to be difficult and more influencedby the measurement errors and hence the static EIT is foundstill far from clinical applications primarilyThough the staticEIT is found difficult due to the fundamental ill-posednessof the EIT inverse problem but still it is little bit too earlyto say that the static EIT is not to be pursued since activeresearchers are still looking for innovative algorithms andnew measurement technologies

The dynamic or difference EIT that was introducedby Barber and Brown in 1984 [174] produces differentialimages whereas the static imaging yields absolute imagesThe difference impedance imaging may be either the timedifference EIT [105 175 176] or frequency difference EIT[105 177 178] The time difference EIT imaging producesthe images of the variation of the conductivity distributionof a region between two different time intervals [104 179]whereas the frequency difference imaging reconstructs theimpedance distribution from two data sets collected at twodiscrete frequencies The difference EIT allows us to monitorthe changes such as gastric emptying or long-term observa-tion of body functionsvolume changes [104 179] including

the visualization of physiological activities in a human bodysuch as respiration It can also be used for breast cancerdetection cardiac circulation brain function stomach emp-tying fracture healing bladder filling and others DifferenceEIT can also be used in nonmedical applications such as incorrosion detection crack detection electric field sensingbubble detection and other nondestructive testing

77Multifrequency EIT System As the different biological tis-sues have different frequency versus impedance responses [576 180 181] the multifrequency EIT [182ndash185] is found moreeffective and efficient in biological tissue characterizationmedical imaging and clinical diagnosis of the human dis-eases The multifrequency EIT provides a lot of informationabout the tissue health which can be potentially utilized forbetter tissue characterization Moreover the multifrequencyEIT can provide all the parameters in the frequency domainwhich adds another advancement in the diagnosis andtreatment of the diseaseThus themultifrequency impedanceimaging can be suitably applied in diagnosis and treatment ofthe human diseases tumors lesions and cancers

78 3D EIT 2D electrical impedance tomography (2D-EIT) reconstructs the approximate spatial distribution of theinternal impedance profile of aDUT from the boundarymea-surements of voltage-current data developed by a constantcurrent injection through the surface electrodes fixed in a2D plane within the patientrsquos body [104] Thus the 2D EITapproximately reconstructs the 2D conductivity distributionby assuming the electrical current conducts in a 2D planeBut as the electrical current is not confined in the planeof electrode array rather it spreads over a 3D space withinthe volume conductor the 2D EIT suffers from the errorscontributed by this 3D conduction of electrical current [104]Moreover 2D EIT of a tissue under test provides only the2D impedance distribution in its tomograms in which the3D anatomical and physiological information is not availableBut 3D EIT provides 3D conductivity distribution with abetter and more scientific visualization of the tissue interiorswhich helps the doctors and clinicians with better tissuecharacterization [104]


79 A Basic EIT System A basic EIT system contains threeparts (Figure 15) EIT instrumentation [115ndash123] a PC withreconstruction algorithm [124ndash133] and an array of EITsensors or surface electrodes [57 102 106 110 186] attachedto a practical phantom or a subject under test (SUT)

In EIT an array of sensors or surface electrodes [57102 106 110 186] is attached to the boundary of the DUTA low frequency sinusoidal current of constant amplitudeis generated by a constant current source and it is injectedthrough the driving electrodes and the boundary potentialsare measured on the sensing electrodes using a particularcurrent injection and voltage measurement fashion calledthe current pattern [100 187 188] Boundary potentials aremeasured using a data acquisition system (DAS) [108 121123] and the processed data obtained from the data processingcircuit containing amplifier and filters (Figure 15) are sent tothe PC for computation and image reconstruction Thus abasic EIT system has three main parts

(i) EIT-instrumentation

(ii) PC with reconstruction algorithm

(iii) electrode array or EIT sensors attached to the subjectunder test (SUT)

710 EIT Instrumentation A common EIT Instrumentationhas four main parts

(i) constant current injector (CCI)

(ii) signal conditioner block (SCB)

(iii) electrode switching module (ESM)

(iv) data acquisition system (DAS)

Constant current injector (CCI) may be developed witha signalfunction generator and a voltage controlled currentsource (VCCS) which is used to inject a constant current sig-nal to the domain boundary through the driving electrodesThe developed voltage signals on the sensing electrodes areprocessed by signal conditioner block (SCB) and sent to theDAS for data acquisition The boundary data are collectedwith a particular current pattern by switching driving andsensing electrode in a particular fashion using a electrodeswitching module (ESM) The boundary data collected froma complete EIT scan of the object are sent to the PC for imagereconstruction

711 Electrode Array or EIT Sensors Surface electrode arrayis a very important part of an EIT system as the boundarydata quality and thus the image quality depends on it As theEIT electrodes are used to inject current to the body undertest and collecting the voltage data at boundary electrodeselectrode parameters are very crucial in EIT imaging Fordesigning a better EIT electrode system one should carefullyselect the electrode parameter such as electrode number elec-trodematerial electrode geometry (shape and size) electrodemodel (point gap shunt and complete) or electrode type(noncompound or compound)

712 Reconstruction Algorithm Image reconstruction algo-rithm is one of the most important parts of EIT system asthe image reconstruction is conducted by the algorithm Theboundary potential data collected from the real objects (SUT)are sent to the PC and are processed to reconstruct the EITimages using the EIT reconstruction algorithm EIT imagereconstruction algorithm has generally two main parts

(i) forward solver (FS)(ii) inverse solver (IS)

The forward solver solves the characteristic equation orgoverning equation of EIT [102ndash110 189ndash194] and computesthe boundary potentials called calculated boundary data (119881119888)for a known constant current simulation [119862] in PC Thecalculated potential data (119881119888) are then compared with themeasured potential data (119881119898) in inverse solver (IS) and theconductivity distribution of DUT is reconstructed for whichthe difference between 119881119898 and 119881119888 (Δ119881 = 119881119898 minus 119881119888) isminimized

TheGauss-Newton based EIT image reconstruction algo-rithm (GN-EIRA) is generally developedwith a finite elementmethod (FEM) [189ndash196] based forward solver (FEM-FS) aGauss-Newton based inverse solvers (GN-IS) in MATLABThe GN-EIRA is developed with Gauss-Newton based mini-mization algorithm (GNMA) and Newton Raphson iterativetechnique (NRIT) [189ndash196] The forward solver generallyapplies a numerical technique like FEM or boundary ele-ment method (BEM) [196] or else and calculates the nodalpotentials [195] within the discretized domain Using theforward solution data the algorithm construct the sensitivitymatrix called Jacobean (119869) which is used by the inverse solverThe inverse solver uses the voltage difference vector (Δ119881)and 119869 and calculates a correction or update in conductivitydistribution to modify it for minimizing the voltage datamismatch Inverse solver repetitively solves the forwardsolution and tries to compute an approximate conductivitydistribution for which the difference between measured andcalculated voltage data is minimized

7121 GN-EIRA In EIT the GN-EIRA is developed with aFEM based forward solver (FS) and Gauss-Newton methodbased inverse solver (GNIS) whereas the GNIS is developedwith Gauss-Newton basedminimization algorithm (GNMA)and Newton Raphson Iterative technique (NRIT) The for-ward solver solves the EIT governing equation by developingthe forward model of the EIT problem and calculates theboundary potentials (119881119888) for a known current injection[119862] applied to the boundary of the domain with a knownconductivity distribution [1205900] The solution obtained fromthe FP is also used to compute the potential mismatchvector [Δ119881 = 119881119898 minus 119881119888] and the Jacobian (119869) that areused in inverse problem to compute the conductivity updatevector [Δ120590] The GNIS solves the inverse problem (IP) usingthe boundary data [119881119888] computed by the forward solverand tries to estimate the domain conductivity distributionfrom measured boundary potential data (119881119898) developedfor the real current injection of the same amplitude TheGNIS in EIT repetitively calculates the conductivity update


Ω

Current injection to the phantom

Constant currentinjector

Voltage signal from phantom

Signalconditioner block

Data acquisitionsystem

Mul

tiple

xer m

odul

e

CPU

Phantom withsensors

1

23

456

7

8

9

1011

12 1314

15

16

Y

X

minus025

0

025

minus005 0 005

Figure 15 Schematic of a basic EIT system with patients with surface electrodes attached to its transthoracic region for thoracic imaging

PC withreconstruction

algorithm

Start

Guess

Stop

CPU

No

Yes

Calculated data (Vc)

12059001205900

120590kJ(120590k) [Vc]k = Γ(120590k)

[120590]k+1 = [120590]k + [Δ120590]k[ΔV]k = [Vc]k minus [Vm]k

Measureddata [Vm]k

Regularizationparameter 120582k

ΔV lt 120576

Y

X

minus025

0

025

minus005 0 005

[Δ120590]k = [JTJ + 120582kI]minus1lfloorJTrfloor[ΔVk]

Figure 16 The flow chart of a standard EIT imaging reconstruction algorithm

[Δ120590] vector and simultaneously reconstructs the conductivitydistribution [120590] in each iteration by updating the conductivitydistribution obtained in previous iteration The iterationprocess is continued to update the conductivity distributionwith modified NRIT until a specified error limit criterion(120576 = 119891(Δ119881)) is obtained (Figure 16)

7122 Governing Equation of EIT The EIT reconstructs theconductivity of a closed domain from the boundary voltagecurrent data using the relationship of the domain conduc-tivity distribution and the domain potentials distribution

which is called the EIT governing equation In EIT theconductivity distribution of a DUT could be defined withthe distinct values of electrical conductivities at each pointswithin the domain which are associated with their corre-sponding coordinates Now if an electric current of constantamplitudes is injected to the domain boundary (120597Ω) thecurrent flux will develop a particular potential profile withinDUT The developed potential profile will be the functionof the current amplitude and the conductivity profile of thedomain Therefore the ldquoEIT governing equationrdquo could beobtained as the relationship between the elemental electrical


conductivity (120590119890) of the domain and their correspondingpotential values (Φ) using the Maxwellrsquos equations EITgoverning equation can be represented as a partial differentialequitation [102ndash110 189ndash195] as shown in

nabla sdot 120590nabla120601 = 0 (14)

7123 Boundary Conditions The forward solver of EITreconstruction algorithm discretizes the DUT with a finiteelement mesh containing a finite number of triangularelements and the finite number of nodes and applies the FEMformulation technique on the EIT governing equation (14)of the DUT As the EIT governing equation is a nonlinearpartial differential equation it has an infinite number ofsolutions and hence to restrict its solutions space the FEMbased EIT forward solution process essentially needs theboundary conditions [1ndash5] which provide some specifiedvalue of certain system parameters which may be either thepotentials at the surface or the current density crossing theboundary or mixed conditions

The boundary conditions in which the parameters arethe potential at the domain boundary (120597Ω) are named as theDirichlet boundary conditionswhich are represented as [102ndash110 189ndash196]

Φ = Φ119894 (15)

where 119894 = 1 119889 are the potentials on the nodes underelectrodes (119889 is number of electrodes potentials)

On the other hand the boundary conditions of EITwhichspecify the current density crossing the boundary (120597Ω) areknown as the Neumann boundary conditions [102ndash110 189ndash196] defined as

int

120597Ω

120590

120597Φ

120597119899

=

+119868 on the source electrode(119868 is the injected current amplitude)

minus119868 on the sink electrode(119868 is the injected current amplitude)

0 otherwise(16)

7124 Forward Solution In EIT the FS derives the forwardmodel of a DUT from its governing equation by applyingthe FEM formulation technique on (1) and imposing theboundary conditions Using the initial guessed conductivitydistribution [1205900] and nodal coordinates the FS develops theforward model which is a system of equations representedin the form of a matrix equation (17) The forward modelin form of the matrix equation actually represents therelationship between the current injectionmatrix [119862] (matrixof the applied current signal) and the nodal potential matrix[Φ] (matrix of the developed nodal potential data) throughthe transformation matrix [119870(120590119890)] as given below [102ndash110189ndash195]

[Φ] = [119870 (120590119890)]minus1[119862] (17)

where 120590119890 is the elemental conductivity distributions

The FS in EIT thus develops the forward model of theDUT and calculates the boundary potential (119881119888) data for aknown conductivity distribution (initial guess) [120590119890 = 1205900]and a known simulated current injection [119862] (available fromthe boundary conditions)

7125 Inverse Solution If 119865 is a function mapping an edimensional (e is the number of element in the FEM mesh)impedance distribution into a set of 119889 (119889 is the numberof the experimental measurement data ([119881119898]) available)approximate measured voltages then the conductivity of theDUT can be estimated from the boundary data by applyingthe Gauss-Newton method based minimization algorithm[102ndash110 189ndash195] which tries to find a least square solutionof the minimized object function 119904119903 [102ndash110 189ndash195] whichis defined as

119904119903 =

1

2

1003817100381710038171003817119881119898 minus 119865

1003817100381710038171003817

2+

1

2

120582 119877120590

=

1

2

(119881119898 minus 119865)119879(119881119898 minus 119865) +

1

2

120582(119877120590)119879(119877120590)2

(18)

where 119865 represents the forward model predicted boundarypotential data or 119881119888 119877 is a matrix which is called the regular-ization matrix operator and 120582 (positive scalar) is called theregularization coefficient or regularization parameters [102ndash110 189ndash195]

Thus the Gauss Newton method based minimizationtechnique of EIT image reconstruction algorithm gives theconductivity update vector [Δ120590] as

Δ120590 =

1199041015840

119903

11990410158401015840119903

=

(1198651015840)

119879

(119881119898 minus 119865) minus 120582(119877)119879(119877120590)

(1198651015840)119879(1198651015840) minus (119865

10158401015840)119879(119881119898 minus 119865) + 120582119877

119879119877

(19)

Now neglecting the higher order matrix [11989110158401015840]119879 in (19)called the ldquoHessian matrixrdquo and replacing the matrix [1198651015840] bythe sensitivity matrix of the system [119869] called the ldquoJacobianmatrixrdquo of the system the conductivity update vector [Δ120590]reduces to

Δ120590 =

119869119879(119881119898 minus 119891) minus 120582119868120590

119869119879119869 + 120582119868

(20)

where the matrix [119868] represents the identity matrix which isequal to [119877119879][119877]

Therefore the GN-EIRA of the EIT provides the generalsolution of the conductivity distribution of the DUT at the119896th iteration of the algorithm as

[120590119896+1] = [120590119896] + [[119869119879119869 + 120582119868]

minus1

[119869119879[119881119898 minus 119891] minus 120582119868120590]]

119896 (21)

713 Subject Under Test (SUT) in EIT The subject to beimaged or subject under test (SUT) in EIT may be a humansubject (healthy human volunteer or a patient) or laboratoryanimal or a tissue mimicking model object called ldquopracticalphantomrdquo [117 129 131 186 188 197ndash202] The EIT instru-ment called electrical impedance tomograph is applied tothe SUT and the EIT imaging is conducted for the domain


of interest or domain under test (DUT) at a particular planewithin the SUT in 2D or for a number of parallel 2D EITdomains called image slices in 3D EIT

7131 Practical Phantoms EIT phantoms [117 129 131 186188 197ndash202] which are the mimic models of the humanbody tissue or human body parts are essentially required toassess the performance of electrical impedance tomography(EIT) systems for their validation calibration and compar-ison purposes EIT phantoms are classified mainly into twotypes namely network phantoms [201 203ndash209] and realobject phantoms [115 186 187 198ndash200 202 210ndash215] Realobject phantoms are developed with a phantoms tank filledwith two or more objects (biological or nonbiological) withdifferent conductivities surrounded by an array of surfaceelectrodes [97] housed on the inner wall of the phantom tankResearchers have reported several types of real object phan-toms for studying their EIT systems such as saline-insulatorphantoms [115 186 187 210ndash215] saline-agar phantom [216ndash218] saline-vegetable phantom [198 212 215 219 220] andpassive or active element phantoms [200 221] All the phan-toms have their own advantages and disadvantages Networkphantoms are developed with circuit elements housed on aelectronic element base or printed circuit board (PCB) Net-work phantoms do not require any surface electrodes onlyelectronic connections are sufficient to interface the phantomwith the EIT instrumentationThough the network phantomswith electrical components are sometime advantageous fortheir long life rigidity stability and ease of control yet thesephantoms fail to provide a mimic model the real biologicaltissue or the real body parts

7132 Disadvantages of Saline Phantoms Saline phantomswith solid inorganic (plastic wood or metal rod) or organic(vegetables or other biological tissues) materials are verypopular in EIT as they are low cost and easy to developand can be developed with rigid or flexible tanks of anyshape and size However they cannot be assumed as a perfectmimic of the body parts as the background medium is apure saline solution (purely resistive) whereas none of thehuman body parts are purely resistive in nature Also itis quite difficult to reconstruct the actual resistivity of theinsulator inhomogeneity in a saline background because oftheir large resistivity difference As the frequency versusimpedance response of the saline solution is constant thesaline phantom fails to present a mimic model of bodytissue as the response of the real tissue varies with frequencyaccording to their physiological and physiochemical compo-sitions contributing to a complex bioelectrical impedanceMoreover the evaporation of the saline solution makesthem unstable over the time which makes the assessmenterroneous in real case Hence the saline phantoms and thephantoms with circuit elements sometimes fail to calibratethemultifrequency EIT systems and themedical EIT systemsIt is found that the practical biological phantoms consistingtwo different materials with low resistivity difference such asthe real tissue phantoms [188 197 200] are more suitable formultifrequency impedance imaging studies

714 Scope andChallenges in EIT Though theEIT technologyhas been studied in different fields of science and technologyyet EIT has shown its several advantages in medical imagingand hence a number of research groups are working onit for medical and clinical imaging purposes Althoughdue to poor signal to noise ratio (SNR) [222ndash225] of theboundary potential data and poor spatial resolution [226ndash228] the EIT technology has a lot of scope to work on itand a number of technical challenges to be solvedThereforethough a medical EIT system has a number of advantagesover the available regular medical imaging modalities andthe technical challenges and research scope of EIT technologyhave ever attracted the researcher to work for improving EITreconstruction A lot of research opportunities are still therein absolute impedance imaging high speed reconstructionimproved 3D reconstruction and spatial resolution of EIT

8 Present Status and Future Directions ofImpedance Methods

Multifrequency bioelectrical impedance analysis (BIA) [229230] is found as one of the core interest among the researchersin the field of bioelectrical impedance A part of theresearchers are studying on the instrumentation who aredeveloping more sophisticated instrumentation The futurestudies may be conducted on the wireless and Bluetoothbased instrumentation for BIA techniques Among the mul-tifrequency impedance analyzing techniques EIS is the mostpopular and strong method EIS methods are now beingstudied with a pulsed signal based EIS instrumentation [231ndash233] for tissue characterizations as well as single cell analysis[231ndash233]The wireless and Bluetooth based instrumentationfor EIS techniques can also be implemented in future studiesPresently the IPG is being studied for finger plethysmography[234 235] by J G Webster group in University of WisconsinMadison USA J G Webster group is also working on theautomated IPG instrumentation and the modern softwarebased IPG systems [236ndash238] In future the wireless andBluetooth based instrumentation for IPG techniques canalso be implemented ICG has several advantages in cardiacparameter assessment over the other conventional invasivemethods In recent years the ICG instruments are availablefrom few industry-institute research collaborations ICG canbe studied as multifrequency bioimpedance methods fortransthoracic parameter assessment for better cardiac healthmonitoring ICG can also be studied as an ambulatorymonitoring or long-term monitoring modality in intensivecare unit (ICU) The electrical behavior of biological tissue isvery complex and hence though the BIA EIS IPG ICG arefound with some successful applications yet the EIT has notyet been considered as the regular medical imagingmodalityNonlinearity ill-posedness modelling error measurementerror and other challenges are still required to be overcomeBut due to some unique advantages if these challengesare overcome by future research on EIT it may be alsoappliedmore effectively and efficiently compared to the otherimaging modalities which are now being commonly usedin some particular medical applications like brain imaging


[239ndash241] breast imaging [242ndash249] abdominal imaging[250ndash252] whole body imaging [253ndash256] and so forthTherefore the EIT technology has a lot of potentials forlow cost fast tomographic imaging though the problemsof low spatial resolution and poor signal to noise ration ofthe system should be solved in future research Absoluteconductivity imaging better electrode performance accuratesystem modelling and high speed 3D EIT are also to beexplored more to improve the EIT technology

9 Conclusions

Bioelectrical impedancemethods used for noninvasive healthmonitoring are reviewed in detail in this paper The paperreviewed the technical aspects of some major bioimpedancemethods such as bioelectrical impedance analysis (BIA) elec-trical impedance spectroscopy (EIS) electrical impedanceplethysmography (IPG) impedance cardiography (ICG) andelectrical impedance tomography (EIT) Detailed discussionson the theoretical aspects working principles applicationsadvantages limitations and present research scenario futuretrends and challenges of these bioimpedance methods havebeen presented in detail in this paper It is found all themethods are comparatively low cost fast portable andsimple All the methods have a number of advantages in theirapplication for the noninvasive investigations compared tothe other methods and hence they have been potentiallyapplied for noninvasive diagnosis of tissue health BIA EISIPG and ICG have been successfully studied by a number ofresearchers and scientists over past few decades for assessinga number of tissue parameters and the tissue compositionsfor tissue health diagnosis BIA EIS IPG ICG and EIT havebeen used for biological tissue characterization by utilizingthe electrical impedance information BIA EIS IPG and ICGstudy the tissue properties in terms of the lump impedanceparameters at a particular frequency or at different discretefrequencies obtained from the boundary voltage currentmeasurement On the other hand the EIT provides a spatialdistribution of the impedance profile of a 2D or 3D domainunder test using a set of boundary voltage-current data Inthis regard EIT provides more information about the tissuephysiology and pathology and hence it has more potential inseveral applications The paper presents a clear and detailedtechnical overview of BIA EIS IPG ICG and EIT methodsand their applications for noninvasive tissue characterizationand tissue health diagnosis which will help the readers to getthe clear technical perspectives of these impedance analyzingtechniques

Conflict of Interests

The author declares that there is no conflict of interestsregarding the publication of this paper

References

[1] F H Netter Atlas of Human Anatomy Rittenhouse BookDistributors Inc 2nd edition 1997

[2] J J Ackmann ldquoComplex bioelectric impedance measurementsystem for the frequency range from 5Hz to 1MHzrdquo Annals ofBiomedical Engineering vol 21 no 2 pp 135ndash146 1993

[3] J J Ackmann andM A Seitz ldquoMethods of complex impedancemeasurements in biologic tissuerdquoCritical Reviews in BiomedicalEngineering vol 11 no 4 pp 281ndash311 1984

[4] K Cha G M Chertow J Gonzalez J M Lazarus and D WWilmore ldquoMultifrequency bioelectrical impedance estimatesthe distribution of body waterrdquo Journal of Applied Physiologyvol 79 no 4 pp 1316ndash1319 1995

[5] T K Bera and J Nagaraju ldquoElectrical impedance spectroscopicstudy of broiler chicken tissues suitable for the development ofpractical phantoms in multifrequency EITrdquo Journal of ElectricalBioimpedance vol 2 pp 48ndash63 2011

[6] A D Bauchot F R Harker and W M Arnold ldquoThe use ofelectrical impedance spectroscopy to assess the physiologicalcondition of kiwifruitrdquo Postharvest Biology and Technology vol18 no 1 pp 9ndash18 2000

[7] E C Hoffer C K Meador and D C Simpson ldquoCorrelation ofwhole-body impedance with total body water volumerdquo Journalof Applied Physiology vol 27 no 4 pp 531ndash534 1969

[8] H C Lukaski P E JohnsonWW Bolonchuk andG I LykkenldquoAssessment of fat-free mass using bioelectrical impedancemeasurements of the human bodyrdquoAmerican Journal of ClinicalNutrition vol 41 no 4 pp 810ndash817 1985

[9] U G Kyle I Bosaeus A D De Lorenzo et al ldquoBioelectricalimpedance analysis-part I review of principles and methodsrdquoClinical Nutrition vol 23 no 5 pp 1430ndash1443 2004

[10] U G Kyle I Bosaeus A D De Lorenzo et al ldquoBioelectricalimpedance analysismdashpart II utilization in clinical practicerdquoClinical Nutrition vol 23 no 6 pp 1430ndash1453 2004

[11] R F Kushner ldquoBioelectrical impedance analysis a review ofprinciples and applicationsrdquo Journal of the American College ofNutrition vol 11 no 2 pp 199ndash209 1992

[12] G Parrinello S Paterna P Di Pasquale et al ldquoThe usefulness ofbioelectrical impedance analysis in differentiating dyspnea dueto decompensated heart failurerdquo Journal of Cardiac Failure vol14 no 8 pp 676ndash686 2008

[13] A P Hills and N M Byrne ldquoBioelectrical impedance and bodycomposition assessmentrdquoMalaysian Journal of Nutrition vol 4pp 107ndash112 1998

[14] H-Y Hu and Y Kato ldquoBody composition assessed by bioelec-trical impedance analysis (BIA) in patients with Gravesdiseasebefore and after treatmentrdquo Endocrine Journal vol 42 no 4 pp545ndash550 1995

[15] R C Janaway S L Percival and A S Wilson ldquoDecompositionof human remainsrdquo inMicrobiology andAging ClinicalManifes-tations L Steven Percival Ed chapter 13 pp 313ndash334 Springer2009

[16] D Bracco D Thiebaud R L Chiolero M Landry P Burck-hardt and Y Schutz ldquoSegmental body composition assessedby bioelectrical impedance analysis and DEXA in humansrdquoJournal of Applied Physiology vol 81 no 6 pp 2580ndash2587 1996

[17] JM Jakicic R RWing andW Lang ldquoBioelectrical impedanceanalysis to assess body composition in obese adult women theeffect of ethnicityrdquo International Journal of Obesity vol 22 no3 pp 243ndash249 1998

[18] L C Danford D A Schoeller and R F Kushner ldquoComparisonof two bioelectrical impedance analysis models for total bodywater measurement in childrenrdquo Annals of Human Biology vol19 no 6 pp 603ndash607 1992


[19] A Bosy-Westphal B Schautz W Later J J Kehayias D Gal-lagher and M J Muller ldquoWhat makes a BIA equation uniqueValidity of eight-electrodemultifrequency BIA to estimate bodycomposition in a healthy adult populationrdquo European Journal ofClinical Nutrition vol 67 supplement 1 pp S14ndashS21 2013

[20] D A Schoeller ldquoBioelectrical impedance analysis What does itmeasurerdquoAnnals of the NewYork Academy of Sciences vol 904pp 159ndash162 2000

[21] W J Rutherford ldquoComparison of bioelectrical impedance andskinfolds with hydrodensitometry in the assessment of bodycomposition in healthy young adultsrdquo Journal of Research vol6 no 2 pp 56ndash60 2011

[22] M E Orazem and B Tribollet Impedance Spectroscopy TheECS Series of Texts and Monographs Wiley-Inter Sc 2008

[23] J R Macdonald ldquoImpedance spectroscopyrdquo Annals of Biomed-ical Engineering vol 20 no 3 pp 289ndash305 1992

[24] T Houssin J Follet A Follet E Dei-Cas and V Senez ldquoLabel-free analysis of water-polluting parasite by electrochemicalimpedance spectroscopyrdquo Biosensors and Bioelectronics vol 25no 5 pp 1122ndash1129 2010

[25] D A Scrymgeour C Highstrete Y-J Lee J W-P Hsu andM Lee ldquoHigh frequency impedance spectroscopy on ZnOnanorod arraysrdquo Journal of Applied Physics vol 107 no 6Article ID 064312 2010

[26] J Nielsen and T Jacobsen ldquoCurrent distribution effects in ACimpedance spectroscopy of electroceramic point contact andthin film model electrodesrdquo Electrochimica Acta vol 55 no 21pp 6248ndash6254 2010

[27] J Wu Y Ben and H-C Chang ldquoParticle detection by electri-cal impedance spectroscopy with asymmetric-polarization ACelectroosmotic trappingrdquoMicrofluidics and Nanofluidics vol 1no 2 pp 161ndash167 2005

[28] H E Ayliffe A B Frazier and R D Rabbitt ldquoElectricimpedance spectroscopy using microchannels with integratedmetal electrodesrdquo Journal of Microelectromechanical Systemsvol 8 no 1 pp 50ndash57 1999

[29] J M Torrents P Juan-Garcıa and A Aguado ldquoElectricalimpedance spectroscopy as a technique for the surveillanceof civil engineering structures considerations on the galvanicinsulation of samplesrdquo Measurement Science and Technologyvol 18 no 7 pp 1958ndash1962 2007

[30] T Repo D H Paine and A G Taylor ldquoElectrical impedancespectroscopy in relation to seed viability and moisture contentin snap bean (Phaseolus vulgaris L)rdquo Seed Science Research vol12 no 1 pp 17ndash29 2002

[31] T Repo J Laukkanen and R Silvennoinen ldquoMeasurement ofthe tree root growth using electrical impedance spectroscopyrdquoSilva Fennica vol 39 no 2 pp 159ndash166 2005

[32] E Barsoukov and J R Macdonald Impedance SpectroscopyTheory Experiment and Applications Wiley-Interscience 2edition 2005

[33] R V Hill J C Jansen and J L Fling ldquoElectrical impedanceplethysmography a critical analysisrdquo Journal of Applied Physi-ology vol 22 no 1 pp 161ndash168 1967

[34] J Nyboer ldquoElectrical impedance plethysmography a physicaland physiologic approach to peripheral vascular studyrdquo Circu-lation vol 2 no 6 pp 811ndash821 1950

[35] R W Griffiths M E Philpot B J Chapman and K A Mun-day ldquoImpedance cardiography non-invasive cardiac outputmeasurement after burn injuryrdquo International Journal of TissueReactions vol 3 no 1 pp 47ndash55 1981

[36] C J Schuster and H P Schuster ldquoApplication of impedancecardiography in critical caremedicinerdquoResuscitation vol 11 no3-4 pp 255ndash274 1984

[37] H H Woltjer H J Bogaard and P M J M de Vries ldquoThetechnique of impedance cardiographyrdquo EuropeanHeart Journalvol 18 no 9 pp 1396ndash1403 1997

[38] J G Webster Electrical Impedance Tomography Adam HilgerSeries of Biomedical Engineering AdamHilger New York NYUSA 1st edition 1990

[39] D S Holder Electrical Impedance Tomography Methods His-tory and Applications Medical Physics and Biomedical Engi-neering Institute of Physics Publishing Ltd 1st edition 2004

[40] H P Schwan ldquoElectrical properties of tissue and cell suspen-sionsrdquo Advances in Biological and Medical Physics vol 5 pp147ndash209 1957

[41] G Oslash Martinsen S Grimnes and H P Schwan ldquoInterfacephenomena and dielectric properties of biological tissuerdquo Ency-clopedia of Surface and Colloid Science vol 20 pp 2643ndash26532002

[42] D Miklavcic N Pavselj and F X Hart Electric Properties ofTissues Wiley Encyclopedia of Biomedical Engineering JohnWiley amp Sons 2006

[43] Internet Article JAWON Medical South Korea httpwwwjawoncokrengtechnologybody-compositionprinciples-of-biaphp

[44] T G Lohman Advances in Body Compositions Human Kinet-ics Champaign Ill USA 1992

[45] Internet Article Bodystat Limited USA httpwwwbodystatcomscience

[46] K T Francis ldquoBody-composition assessment using underwaterweighing techniquesrdquo Physical Therapy vol 70 no 10 pp 657ndash663 1990

[47] M K Oates ldquoThe Use of DXA for Total Body CompositionAnalysis ndash Part I International Society for Clinical Densitome-tryrdquo SCAN Newsletter vol 13 Quarter 2 6-7

[48] L B Houtkooper T G Lohman S B Going and W HHowell ldquoWhy bioelectrical impedance analysis should be usedfor estimating adiposityrdquoAmerican Journal of Clinical Nutritionvol 64 no 3 supplement pp 436Sndash448S 1996

[49] ldquoInternet Articlerdquo The Body Remedy Health Media New YorkUSA httpwwwthebodycomcontentart14236html

[50] Internet Article and Doylestown Hospital Pennsylvania 18901USA httpwwwdhorgbodycfmid=1023

[51] D N Lobo Z Stanga J A D Simpson J A Anderson BJ Rowlands and S P Allison ldquoDilution and redistributioneffects of rapid 2-litre infusions of 09 (wv) saline and5 (wv) dextrose on haematological parameters and serumbiochemistry in normal subjects a double-blind crossoverstudyrdquo Clinical Science vol 101 no 2 pp 173ndash179 2001

[52] E Barsoukov and J R Macdonald Impedance SpectroscopyTheory Experiment andApplicationsWiley-Interscience 2005


[54] E Mark Orazem Bernard Tribollet Electrochemical ImpedanceSpectroscopy The ECS Series of Texts and Monographs Wiley-Interscience 1 edition 2008

[55] T K Bera Y Mohamadou K H Lee et al ldquoElectricalimpedance spectroscopy for electro-mechanical characteriza-tion of conductive fabricsrdquo Sensors vol 14 no 6 pp 9738ndash97542014


[56] A Lasia ldquoElectrochemical impedance spectroscopy and itsapplicationsrdquo in Modern Aspects of Electrochemistry B EConway J Bockris and R E White Eds vol 32 pp 143ndash248Kluwer AcademicPlenum Publishers New York NY USA1999

[57] B-Y Chang and S-M Park ldquoElectrochemical impedance spec-troscopyrdquo Annual Review of Analytical Chemistry vol 3 no 1pp 207ndash229 2010

[58] H Shih and T-C Lo ldquoElectrochemical impedance spec-troscopy for battery research and developmentrdquo Tech Rep 31Solartron Instruments USA

[59] A Ehsani M G Mahjani and M Jafarian ldquoElectrochemicalimpedance spectroscopy study on intercalation and anomalousdiffusion of AlCl-4 ions into graphite in basic molten saltrdquoTurkish Journal of Chemistry vol 35 no 5 pp 735ndash743 2011

[60] J L Jespersen A E Toslashnnesen K Noslashrregaard L Overgaardand F Elefsen ldquoCapacitymeasurements of li-Ion batteries usingAC impedance spectroscopyrdquo World Electric Vehicle Journalvol 3 2009

[61] M Nahvi and B S Hoyle ldquoElectrical impedance spectroscopysensing for industrial processesrdquo IEEE Sensors Journal vol 9no 12 pp 1808ndash1816 2009

[62] D Loveday P Peterson and B Rodgers ldquoEvaluation of organiccoatings with electrochemical impedance spectroscopyrdquo JCTCoatingsTech vol 2 no 2 pp 22ndash27 2005

[63] M Ates and A S Sarac ldquoElectrochemical impedance spec-troscopy of poly[carbazole-co-N-p-tolylsulfonyl pyrrole] oncarbon fiber microelectrodes equivalent circuits for mod-ellingrdquo Progress in Organic Coatings vol 65 no 2 pp 281ndash2872009

[64] J Yu and C-C Liu ldquoMicrofabricated thin film impedancesensor amp AC impedance measurementsrdquo Sensors vol 10 no 6pp 5845ndash5858 2010

[65] M Metikos-Hukovic and S Omanovic ldquoThin indium oxidefilm formation and growth impedance spectroscopy and cyclicvoltammetry investigationsrdquo Journal of Electroanalytical Chem-istry vol 455 no 1-2 pp 181ndash189 1998

[66] M-G Olivier and M Poelman ldquoUse of electrochemicalimpedance spectroscopy (EIS) for the evaluation of electrocoat-ings performances recent researchesrdquo in Corrosion Evaluationand Protection R S Razavi Ed InTech 2012

[67] F Kurniawan New analytical applications of gold nanoparticles[PhD thesis] University of Regensburg Regensburg Germany2008

[68] J P Lynch K J Loh and N Kotov ldquoElectrical impedanceanalysis of carbon nanotube-polyelectrolyte thin film strainsensorsrdquo in Proceedings of the US-Korea Workshop on SmartStructures Technology for Steel Structures Seoul Korea 2006

[69] A S Sarac M Ates and B Kilic ldquoElectrochemical impedancespectroscopic study of polyaniline on platinum glassy carbonand carbon fiber microelectrodesrdquo International Journal ofElectrochemical Science vol 3 pp 777ndash786 2008

[70] W Yang J E Butler J N Russell Jr and R J Hamers ldquoDirectelectrical detection of antigen-antibody binding on diamondand silicon substrates using electrical impedance spectroscopyrdquoAnalyst vol 132 no 4 pp 296ndash306 2007

[71] A Vandenberg and K J Loh ldquoEvaluating the Ph sensitivity ofcarbon nanotube-polyaniline thin films with different dopantsrdquoNano LIFE vol 2 no 4 Article ID 1242001 12 pages 2012

[72] D A Dean T Ramanathan D Machado and R SundararajanldquoElectrical impedance spectroscopy study of biological tissuesrdquoJournal of Electrostatics vol 66 no 3-4 pp 165ndash177 2008

[73] P Heroux and M Bourdages ldquoMonitoring living tissues byelectrical impedance spectroscopyrdquo Annals of Biomedical Engi-neering vol 22 no 3 pp 328ndash337 1994

[74] J Estrela da Silva J P Marques de Sa and J Jossinet ldquoClassi-fication of breast tissue by electrical impedance spectroscopyrdquoMedical and Biological Engineering and Computing vol 38 no1 pp 26ndash30 2000

[75] C Skourou P J Hoopes R R Strawbridge and K D PaulsenldquoFeasibility studies of electrical impedance spectroscopy forearly tumor detection in ratsrdquo Physiological Measurement vol25 no 1 pp 335ndash346 2004

[76] T K Bera J K Seo H Kwon and J Nagaraju ldquoA LabVIEWbased electrical bio-impedance spectroscopic data interpreter(LEBISDI) for studying the equivalent circuit parameters ofbiological tissuesrdquo in Proceedings of the 15th InternationalConference on Electrical Bio-Impedance (ICEBI) and 14th Con-ference on Electrical Impedance Tomography (EIT) p 77 BerlinGermany April 2013

[77] J M Cruza I C Fitaa L Sorianob J Payab and M VBorracherob ldquoThe use of electrical impedance spectroscopy formonitoring the hydration products of Portland cement mortarswith high percentage of pozzolansrdquo Cement and ConcreteResearch vol 50 pp 51ndash61 2013

[78] L Dhouibi E Triki and A Raharinaivo ldquoThe applicationof electrochemical impedance spectroscopy to determine thelong-term effectiveness of corrosion inhibitors for steel inconcreterdquo Cement and Concrete Composites vol 24 no 1 pp35ndash43 2002

[79] A Akhavan and F Rajabipour ldquoEvaluating ion diffusivity ofcracked cement paste using electrical impedance spectroscopyrdquoMaterials and Structures vol 46 no 5 pp 697ndash708 2013

[80] M Tiitta andH Olkkonen ldquoElectrical impedance spectroscopydevice for measurement of moisture gradients in woodrdquo Reviewof Scientific Instruments vol 73 no 8 p 3093 2002

[81] M Tiitta T Savolainen H Olkkonen and T Kanko ldquoWoodmoisture gradient analysis by electrical impedance spec-troscopyrdquo Holzforschung vol 53 no 1 pp 68ndash76 1999

[82] S L Zelinka D R Rammer and D S Stone ldquoImpedancespectroscopy and circuit modeling of Southern pine above 20moisture contentrdquo Holzforschung vol 62 no 6 pp 737ndash7442008

[83] X Liu Electrical impedance spectroscopy applied in plantphysiology studies [MS thesis] RMIT University MelbourneAustralia 2006

[84] A Hakam M M Takam M Chokairi et al ldquoEffect of barkstripping on the electrical impedance of quercus suber leavesmaderasrdquo Ciencia y Tecnologıa vol 14 no 2 pp 195ndash208 2012

[85] S Ha M Diez-Silva E Du et al ldquoMicrofluidic electricimpedance spectroscopy for Malaria diagnosisrdquo in Proceedingsof the 16th International Conference on Miniaturized Systemsfor Chemistry and Life Sciences pp 1960ndash1962 Okinawa Japan2012

[86] S Ha A malaria diagnostic system based on electric impedancespectroscopy [MS thesis] Massachusetts Institute of Technol-ogy Boston Mass USA 2011


[88] A Fattah-alhosseini S Taheri Shoja B Heydari Zebardastand P Mohamadian Samim ldquoAn electrochemical impedance


spectroscopic study of the passive state on AISI 304 stainlesssteelrdquo International Journal of Electrochemistry vol 2011 ArticleID 152143 8 pages 2011

[89] A M A Alsamuraee H I Jaafer H A Ameen and A Q Ab-dullah ldquoElectrochemical impedance spectroscopic evaluationof corrosion protection properties of polyurethanepolyvinylchloride blend coatings on steelrdquo American Journal of Scientificand Industrial Research vol 2 no 5 pp 761ndash768 2011

[90] T K Bera and J Nagaraju ldquoElectrical impedance spectroscopicstudies of the electronic connectors of DIP switch basedmultiplexers suitable for multifrequency electrical impedancetomography biomedical engineering Narosa PublishingHouserdquo in Proceeding of the International Conference onBiomedical Engineering (ICBME rsquo11) pp 58ndash65 Manipal India2011

[91] PM Gomadam and JWWeidner ldquoAnalysis of electrochemicalimpedance spectroscopy in proton exchange membrane fuelcellsrdquo International Journal of Energy Research vol 29 no 12pp 1133ndash1151 2005


[93] T Ryll A Brunner S Ellenbroek A Bieberle-Hutter J L MRupp and L J Gauckler ldquoElectrical conductivity and crystal-lization of amorphous bismuth ruthenate thin films depositedby spray pyrolysisrdquo Physical Chemistry Chemical Physics vol 12no 42 pp 13933ndash13942 2010

[94] J Nyboer ldquoPlethysmography Impedancerdquo in Medical PhysicsOGlasser Ed vol 2 pp 736ndash743 Year Book Chicago Ill USA1950

[95] J Nyboer S Bango and L F Nims ldquoThe impedance plethysmo-graph and electrical volume recorderrdquo CAM Report 149 OSPR1943

[96] D K Swanson and J G Webster ldquoOrigin of the electricalimpedance pulse in the limbsrdquo in Proceedings of the 29th AnnualConference on Engineering in Medicine amp Biology vol 18 p 3241976

[97] J G Webster Medical Instrumentation Application and DesignWiley 4 edition 2009

[98] D Wade Peterson ldquoInternet Article Phoenix Ambula-tory Blood Pressure Monitor Project Sub-project Impe-dance Plethysmograph Sensorrdquo httpwwwthebodycomcon-tentart14236html

[99] Internet Article httpwwwstinasagovttospinoff2001hm1html

[100] J Malmivuo and R Plonsey Bioelectromagnetism Principlesand Applications of Bioelectric and Biomagnetic Fields OxfordUniversity Press New York NY USA 1995

[101] Internet Article ldquoOverview of Impedance Cardiography(ICG)rdquo httpimpedancecardiographycomicgover10html

[102] M H F Ribeiro R W dos Santos L Paulo S Barra and FC Peters ldquoSimulation study on the determination of cardiacejection fraction by electrical impedance tomography usinga hybrid heuristic approachrdquo Journal of Medical Imaging andHealth Informatics vol 4 pp 113ndash121 2014

[103] T K Bera ldquoNoninvasive electromagnetic methods for brainmonitoring a technical reviewrdquo in Brain Computer InterfacesCurrent Trends and Applications Intelligent Systems ReferenceLibrary Chapter 2 Springer 2014

[104] T K Bera and J Nagaraju ldquoElectrical Impedance Tomography(EIT) a harmlessmedical imagingmodalityrdquo inResearchDevel-opments in Computer Vision and Image Processing Methodolo-gies and Applications chapter 13 pp 224ndash262 IGI Global USA

[105] J K Seo and E J Woo Nonlinear Inverse Problems in ImagingWiley 1 edition December 2012

[106] T K Bera and J Nagaraju ldquoSensors for electrical impedancetomographyrdquo in The Measurement Instrumentation and Sen-sors Handbook J G Webster Ed chapter 61 pp 611ndash6130CRC Press 2nd edition 2014

[107] PMetherallThree dimensional electrical impedance tomographyof the human thorax [PhD thesis] University of Sheffield 1998

[108] T K Bera Studies on multi frequency Electrical ImpedanceTomography (EIT) to improve the impedance imaging forbiomedical applications [PhD thesis] Department of Instru-mentation and Applied Physics Indian Institute of ScienceBangalore India

[109] P Metherall D C Barber R H Smallwood and B H BrownldquoThree-dimensional electrical impedance tomographyrdquoNaturevol 380 no 6574 pp 509ndash512 1996

[110] T K Bera and J Nagaraju ldquoStudies on thin film based flex-ible gold electrode arrays for resistivity imaging in electricalimpedance tomographyrdquo Measurement vol 47 pp 264ndash2862014

[111] L Borcea ldquoElectrical impedance tomographyrdquo Inverse Prob-lems vol 18 no 6 pp R99ndashR136 2002

[112] B H Brown ldquoElectrical impedance tomography (EIT) areviewrdquo Journal of Medical Engineering and Technology vol 27no 3 pp 97ndash108 2003

[113] R H Bayford ldquoBioimpedance tomography (electricalimpedance tomography)rdquo Annual Review of BiomedicalEngineering vol 8 pp 63ndash91 2006

[114] E L Costa R G Lima andM B Amato ldquoElectrical impedancetomographyrdquo Current Opinion in Critical Care vol 15 no 1 pp18ndash24 2009

[115] T K Bera and N Jampana ldquoA Multifrequency constant cur-rent source for medical electrical impedance tomographyrdquo inProceedings of the IEEE International Conference on Systems inMedicine and Biology pp 290ndash295 Kharagpur India Decem-ber 2010

[116] T I Oh H Koo K H Lee et al ldquoValidation of a multi-frequency electrical impedance tomography (mfEIT) systemKHU Mark1 impedance spectroscopy and time-differenceimagingrdquo PhysiologicalMeasurement vol 29 no 3 pp 295ndash3072008

[117] T K Bera and J Nagaraju ldquoA study of practical biological phan-toms with simple instrumentation for Electrical ImpedanceTomography (EIT)rdquo in Proceedings of the IEEE Intrumentationand Measurement Technology Conference (I2MTC rsquo09) pp 511ndash516 Singapore May 2009

[118] P J Riu J Rosell A Lozano and R Pallas-Areny ldquoMulti-frequency static imaging in electrical impedance tomographypart 1 instrumentation requirementsrdquo Medical and BiologicalEngineering and Computing vol 33 no 6 pp 784ndash792 1995

[119] T K BeraM Saikia and JNagaraju ldquoABattery-basedConstantCurrent Source (Bb-CCS) for Biomedical Applicationsrdquo inInternational Conference on Computing Communication andNetworking Technologies (ICCCNT rsquo13) Tamilnadu India 2013

[120] T I Oh H Wi D Y Kim P J Yoo and E J Woo ldquoAfully parallelmulti-frequency EIT systemwith flexible electrodeconfiguration KHU Mark2rdquo Physiological Measurement vol32 no 7 pp 835ndash849 2011


[121] T K Bera and J Nagaraju ldquoA LabVIEW based multifunctionmultifrequency electrical impedance tomography (MfMf-EIT)instrumentation for flexible and versatile impedance imagingrdquoin Proceedings of the 15th International Conference on ElectricalBio-Impedance (ICEBI rsquo13) and 14th Conference on ElectricalImpedance Tomography (EIT rsquo13) p 216 Berlin Germany 2013

[122] K G Boone and D S Holder ldquoCurrent approaches to analogueinstrumentation design in electrical impedance tomographyrdquoPhysiological Measurement vol 17 no 4 pp 229ndash247 1996

[123] T K Bera and J Nagaraju ldquoA Battery Based MultifrequencyElectrical Impedance Tomography (BbMf-EIT) system forimpedance imaging of human anatomyrdquo in Proceedings ofthe 15th International Conference on Electrical Bio-Impedance(ICEBI) and 14th Conference on Electrical Impedance Tomogra-phy (EITrsquo 13) p 217 Berlin Germany 2013

[124] W R B Lionheart ldquoEIT reconstruction algorithms pitfallschallenges and recent developmentsrdquo Physiological Measure-ment vol 25 no 1 pp 125ndash142 2004

[125] S Teniou andM Meribout ldquoA new hierarchical reconstructionalgorithm for electrical capacitance tomography using a relax-ation region-based approachrdquo Measurement vol 45 no 4 pp683ndash690 2012

[126] L-M Wang Y-L Zhao F-L Chen and Y Han ldquoThe 3DCT reconstruction algorithm to directly reconstruct multi-characteristic based on EMDrdquoMeasurement vol 44 no 10 pp2043ndash2048 2011

[127] S-W Chen and C-Y Chu ldquoA comparison of 3D cone-beamComputed Tomography (CT) image reconstruction perfor-mance on homogeneous multi-core processor and on otherprocessorsrdquoMeasurement vol 44 no 10 pp 2035ndash2042 2011

[128] T K Bera S K Biswas K Rajan and J Nagaraju ldquoImprovingimage quality in Electrical Impedance Tomography (EIT)using Projection Error Propagation-Based Regularization(PEPR) technique a simulation studyrdquo Journal of ElectricalBioimpedance vol 2 pp 2ndash12 2011

[129] T K Bera S K Biswas K Rajan and J Nagaraju ldquoImagereconstruction in Electrical Impedance Tomography (EIT)withprojection error propagation-based regularization (PEPR) apractical phantom studyrdquo in Advanced Computing Networkingand Security vol 7135 of Lecture Notes in Computer ScienceSpringer 2012

[130] T K Bera S K Biswas K Rajan and J Nagaraju ldquoImprovingconductivity image quality using Block Matrix-based MultipleRegularization (BMMR) technique in EIT a simulation studyrdquoJournal of Electrical Bioimpedance vol 2 pp 33ndash47 2011

[131] T K Bera S K Biswas K Rajan and N Jampana ldquoImprovingthe image reconstruction in Electrical Impedance Tomogra-phy (EIT) with block matrix-based Multiple Regularization(BMMR) a practical phantom studyrdquo in Proceedings of theWorld Congress on Information and Communication Technolo-gies (WICT rsquo11) pp 1346ndash1351 University of Mumbai MumbaiIndia December 2011

[132] F Yang J Zhang and R Patterson ldquoDevelopment of ananatomically realistic forward solver for thoracic electricalimpedance tomographyrdquo Journal of Medical Engineering vol2013 Article ID 983938 7 pages 2013

[133] L Jing S Liu L Zhihong and S Meng ldquoAn image reconstruc-tion algorithm based on the extended Tikhonov regularizationmethod for electrical capacitance tomographyrdquo Measurementvol 42 no 3 pp 368ndash376 2009

[134] J Davis and PWells ldquoComputed tomographymeasurements onwoodrdquo Industrial Metrology vol 2 no 3-4 pp 195ndash218 1992

[135] J Hiller and L M Reindl ldquoA computer simulation platformfor the estimation ofmeasurement uncertainties in dimensionalX-ray computed tomographyrdquo Measurement vol 45 no 8 pp2166ndash2182 2012

[136] G Andria F Attivissimo and A M L Lanzolla ldquoA statisticalapproach for MR and CT images comparisonrdquo Measurementvol 46 pp 57ndash65 2013

[137] X Zhang N Homma S Goto et al ldquoA hybrid image filteringmethod for computer-aided detection of microcalcificationclusters in mammogramsrdquo Journal of Medical Engineering vol2013 Article ID 615254 8 pages 2013

[138] G Betta D Capriglione and N Pasquino ldquoExperimentalinvestigation on workersexposure to electromagnetic fields inproximity of magnetic resonance imaging systemsrdquo Measure-ment vol 45 no 2 pp 199ndash206 2012

[139] B Leporq S Camarasu-Pop E E Davila-Serrano F Pilleuland O Beuf ldquoEnabling 3D-liver perfusion mapping from MR-DCE imaging using distributed computingrdquo Journal of MedicalEngineering vol 2013 Article ID 471682 7 pages 2013

[140] M Almoudi and Z Sun ldquoMyocardial perfusion imaging using99mTc-MIBI single photon emission computed tomography acardiac phantom studyrdquo Journal of Medical Imaging and HealthInformatics vol 3 pp 480ndash486 2013

[141] C Imperiale and A Imperiale ldquoSome fast calculations simu-lating measurements from single-photon emission computedtomography (SPECT) imagingrdquoMeasurement vol 37 no 3 pp218ndash240 2005

[142] L Garcıa-Rojas Adame-Ocampo G Mendoza-Vazquez EAlexanderson and J L Tovilla-Canales ldquoOrbital positron emis-sion tomographycomputed tomography (PETCT) imagingfindings in graves ophthalmopathyrdquo BMC Research Notes vol6 p 353 2013

[143] C A Yi K S Lee H Y Lee et al ldquoCoregistered whole bodymagnetic resonance imaging-positron emission tomography(MRI-PET) versus PET-computed tomography plus brain MRIin staging resectable lung cancer comparisons of clinicaleffectiveness in a randomized trialrdquo Cancer vol 119 no 10 pp1784ndash1791 2013

[144] A Fenster K Surry W Smith and D B Downey ldquoThe useof three-dimensional ultrasound imaging in breast biopsy andprostate therapyrdquo Measurement vol 36 no 3-4 pp 245ndash2562004

[145] G Andria F Attivissimo G Cavone N Giaquinto and AM L Lanzolla ldquoLinear filtering of 2-D wavelet coefficients fordenoising ultrasound medical imagesrdquo Measurement vol 45no 7 pp 1792ndash1800 2012

[146] D S Holder Clinical and Physiological Applications of ElectricalImpedance Tomography Taylor amp Francis UCL Press UCLLondon UK 1st edition 1993

[147] Y Li L Rao R He et al ldquoA novel combination method ofelectrical impedance tomography inverse problem for brainimagingrdquo IEEE Transactions on Magnetics vol 41 no 5 pp1848ndash1851 2005

[148] D Murphy P Burton R Coombs L Tarassenko and P RolfeldquoImpedance imaging in the newbornrdquo Clinical Physics andPhysiological Measurement vol 8 no 4 supplement pp 131ndash140 1987

[149] H A Tyna and S E Iles ldquoTechnology review the use ofelectrical impedance scanning in the detection of breast cancerrdquoBreast Cancer Research vol 6 no 2 pp 69ndash74 2004

[150] F S Moura J C C Aya A T Fleury M B P Amato and R GLima ldquoDynamic imaging in electrical impedance tomography


of the human chest with online transitionmatrix identificationrdquoIEEE Transactions on Bio-Medical Engineering vol 57 no 2 pp422ndash431 2010

[151] F Ferraioli A Formisano and R Martone ldquoEffective exploita-tion of prior information in electrical impedance tomographyfor thermal monitoring of hyperthermia treatmentsrdquo IEEETransactions on Magnetics vol 45 no 3 pp 1554ndash1557 2009

[152] W He P Ran Z Xu B Li and S-N Li ldquoA 3D visualizationmethod for bladder filling examination based on EIT com-putational and mathematical methods in medicinerdquo vol 2012Article ID 528096 9 pages 2012

[153] J L Davidson L S Ruffino D R Stephenson R MannB D Grieve and T A York ldquoThree-dimensional electricalimpedance tomography applied to a metal-walled filtration testplatformrdquo Measurement Science and Technology vol 15 no 11pp 2263ndash2274 2004

[154] G T Bolton C H Qiu and M Wang ldquoA novel electricaltomography sensor for monitoring the phase distribution inindustrial reactorsrdquo in Proceedings of the 7th UK Conference-Fluid Mixing Bradford UK July 2002

[155] F Dickin and M Wang ldquoElectrical resistance tomography forprocess applicationsrdquoMeasurement Science and Technology vol7 no 3 pp 247ndash260 1996

[156] T-C Hou K J Loh and J P Lynch ldquoSpatial conductivitymapping of carbon nanotube composite thin films by electricalimpedance tomography for sensing applicationsrdquoNanotechnol-ogy vol 18 no 31 Article ID 315501 2007

[157] T Sun S Tsuda K-P Zauner and H Morgan ldquoOn-chipelectrical impedance tomography for imaging biological cellsrdquoBiosensors and Bioelectronics vol 25 no 5 pp 1109ndash1115 2010

[158] P Linderholm L Marescot M H Loke and P Renaud ldquoCellculture imaging using microimpedance tomographyrdquo IEEETransactions on Biomedical Engineering vol 55 no 1 pp 138ndash146 2008

[159] F Djamdji A C Gorvin I L Freeston R C Tozer I C Mayesand S R Blight ldquoElectrical impedance tomography appliedto semiconductor wafer characterizationrdquoMeasurement Scienceand Technology vol 7 no 3 pp 391ndash395 1996

[160] K Karhunen A Seppanen A Lehikoinen P J M Monteiroand J P Kaipio ldquoElectrical Resistance Tomography imaging ofconcreterdquo Cement and Concrete Research vol 40 no 1 pp 137ndash145 2010

[161] J E Chambers P B Wilkinson D Wardrop et al ldquoBedrockdetection beneath river terrace deposits using three-dimensional electrical resistivity tomographyrdquo Geomorphologyvol 177-178 pp 17ndash25 2012

[162] P Church J E McFee S Gagnon and P Wort ldquoElectricalimpedance tomographic imaging of buried landminesrdquo IEEETransactions on Geoscience and Remote Sensing vol 44 no 9pp 2407ndash2420 2006

[163] B Ullrich T Gunther and C Rucker ldquoElectrical resistiv-ity tomography methods for archaeological prospectionrdquo inProceedings of the 35th International Conference on ComputerApplications and Quantitative Methods in Archaeology (CAArsquo07) A Posluschny K Lambers and I Herzog Eds pp 1ndash7Berlin Germany 2007 Koll Vor- u Fruhgesch 10 (Bonn 2008)

[164] M Ingham D Pringle andH Eicken ldquoCross-borehole resistiv-ity tomography of sea icerdquoCold Regions Science and Technologyvol 52 no 3 pp 263ndash277 2008

[165] G DAntona and L Rocca ldquoElectrical impedance tomogra-phy for underground pollutant detection and polluted lands

reclaiming monitoringrdquo in Proceedings of the 19th IEEE Intru-mentation and Measurement Technology Conference pp 1035ndash1038 Anchorage Alaska USA May 2002

[166] A Yao C L Yang J K Seo andM Soleimani ldquoEIT-based fabricpressure sensing computational and mathematical methodsin medicinerdquo Computational and Mathematical Methods inMedicine vol 2013 Article ID 405325 9 pages 2013

[167] H Gao C Xu F Fu and S Wang ldquoEffects of particle chargingon electrical capacitance tomography systemrdquo Measurementvol 45 no 3 pp 375ndash383 2012

[168] Z Fan R X Gao and J Wang ldquoVirtual instrument for onlineelectrical capacitance tomographyrdquo in Practical Applicationsand Solutions Using LabVIEW F Silviu Ed under CC BY-NC-SA 30 license 2011

[169] A Perrone G Zeni S Piscitelli et al ldquoJoint analysis of SARinterferometry and electrical resistivity tomography surveys forinvestigating ground deformation the case-study of Satriano diLucania (Potenza Italy)rdquo Engineering Geology vol 88 no 3-4pp 260ndash273 2006

[170] A Finizola A Revil E Rizzo et al ldquoHydrogeological insights atStromboli volcano (Italy) from geoelectrical temperature andCO2 soil degassing investigationsrdquoGeophysical Research Lettersvol 33 no 17 Article ID L17304 2006

[171] A Beauvais M Ritz J-C Parisot C Bantsimba and MDukhan ldquoCombined ERT and GPR methods for investigatingtwo-stepped lateritic weathering systemsrdquo Geoderma vol 119no 1-2 pp 121ndash132 2004

[172] A Kemna B Kulessa and H Vereecken ldquoImaging andcharacterisation of subsurface solute transport using electricalresistivity tomography (ERT) and equivalent transportmodelsrdquoJournal of Hydrology vol 267 no 3-4 pp 125ndash146 2002

[173] H Karabulut S Ozalaybey T Taymaz M Aktar O Selviand A Kocaoglu ldquoA tomographic image of the shallow crustalstructure in the EasternMarmarardquoGeophysical Research Lettersvol 30 no 24 pp 1ndash10 2003

[174] D C Barber and B H Brown ldquoApplied potential tomographyrdquoJournal of Physics E vol 17 pp 723ndash734 1984

[175] J Kuen E J Woo and J K Seo ldquoMulti-frequency time-difference complex conductivity imaging of canine and humanlungs using theKHUMark1 EIT systemrdquoPhysiologicalMeasure-ment vol 30 no 6 pp S149ndashS164 2009


[177] B Harrach J K Seo and E J Woo ldquoFactorization methodand its physical justification in frequency-difference electricalimpedance tomographyrdquo IEEE Transactions on Medical Imag-ing vol 29 no 11 pp 1918ndash1926 2010

[178] J K Seo J Lee S W Kim H Zribi and E J Woo ldquoFrequency-difference electrical impedance tomography (fdEIT) algorithmdevelopment and feasibility studyrdquo Physiological Measurementvol 29 no 8 pp 929ndash944 2008

[179] M Molinary High fidelity imaging in electrical impedancetomography [PhD thesis] University of SouthamptonSouthampton UK 2003

[180] T K Bera and J Nagaraju ldquoStudying the variations of complexelectrical bio-impedance of vegetables and fruits under thedifferent health statusrdquo in Proceedings of the 15th InternationalConference on Electrical Bio-Impedance (ICEBI rsquo13) and 14th


Conference on Electrical Impedance Tomography (EIT rsquo13) p 193Germany April 2013

[181] P L M Cox-Reijven B Van Kreel and P B Soeters ldquoBio-electrical impedance spectroscopy alternatives for the conven-tional hand-to-foot measurementsrdquo Clinical Nutrition vol 21no 2 pp 127ndash133 2002

[182] A Romsauerova A McEwan L Horesh R Yerworth RH Bayford and D S Holder ldquoMulti-frequency electricalimpedance tomography (EIT) of the adult human head initialfindings in brain tumours arteriovenous malformations andchronic stroke development of an analysis method and calibra-tionrdquo Physiological Measurement vol 27 no 5 pp S147ndashS1612006

[183] N K Soni A Hartov C Kogel S P Poplack and K DPaulsen ldquoMulti-frequency electrical impedance tomography ofthe breast new clinical resultsrdquo Physiological Measurement vol25 no 1 pp 301ndash314 2004

[184] H Wi T E Kim T I Oh and E J Woo ldquoExpandable multi-frequency EIT system for clinical applicationsrdquo in Progress InElectromagnetics Research Symposium Proceedings pp 49ndash52Kuala Lumpur Malaysia 2012


[186] T K Bera and J Nagaraju ldquoGold electrode sensors for ElectricalImpedance Tomography (EIT) studiesrdquo in IEEE Sensors Appli-cation Symposium pp 24ndash28 San Antonio Tex USA 2011

[187] T K Bera and J Nagaraju Studying the 2D Resistivity Recon-struction of Stainless Steel Electrode Phantoms Using DifferentCurrent Patterns of Electrical Impedance Tomography (EIT)Biomedical Engineering Narosa Publishing House 2011

[188] T K Bera and J Nagaraju ldquoStudying the resistivity imagingof chicken tissue phantoms with different current patterns inElectrical Impedance Tomography (EIT)rdquo Measurement vol45 no 4 pp 663ndash682 2012

[189] T J Yorkey Comparing reconstruction methods for electricalimpedance tomography [PhD thesis] University of Wisconsinat Madison Madison Wis USA 1986

[190] T J Yorkey J G Webster and W J Tompkins ldquoComparingreconstruction algorithms for electrical impedance tomogra-phyrdquo IEEE Transactions on Biomedical Engineering vol 34 no11 pp 843ndash852 1987


[192] V Marko Electrical Impedance Tomography and Prior Informa-tion KuopioUniversity Publications Natural and Environmen-tal Sciences 1997

[193] C J GrootveldMeasuring and modeling of concentrated settlingsuspensions using electrical impedance tomography [PhD the-sis] Delft University of Technology The Netherlands 1996

[194] B M Graham Enhancements in Electrical Impedance Tomog-raphy (EIT) image reconstruction for 3D lung imaging [PhDthesis] University of Ottawa 2007

[195] T K Bera and J Nagaraju ldquoA FEM-based forward solver forstudying the forward problem of electrical impedance tomog-raphy (EIT) with a practical biological phantomrdquo in Proceedingsof the IEEE International Advance Computing Conference (IACCrsquo09) pp 1375ndash1381 Patiala India March 2009

[196] J N Reddy Introduction to the Finite Element Method TATAMcGraw-Hill Publishing Company Ltd Delhi India 3rd edi-tion 2006

[197] T K Bera and J Nagaraju ldquoA gold sensors array for imagingthe real tissue phantom in electrical impedance tomographyrdquo inInternational Conference on Materials Science and TechnologyKerala India 2012

[198] D SHolder YHanquan andARao ldquoSomepractical biologicalphantoms for calibrating multifrequency electrical impedancetomographyrdquo Physiological Measurement vol 17 no 4 ppA167ndashA177 1996

[199] T K Bera and J Nagaraju ldquoA stainless steel electrode phantomto study the forward problem of Electrical Impedance Tomog-raphy (EIT)rdquo Sensors and Transducers Journal vol 104 no 5pp 33ndash40 2009

[200] T K Bera and J Nagaraju ldquoA chicken tissue phantom forstudying an electrical impedance tomography (EIT) systemsuitable for clinical imagingrdquo Sensing and Imaging vol 12 no3-4 pp 95ndash116 2011

[201] H Griffiths ldquoA phantom for electrical impedance tomographyrdquoClinical Physics and PhysiologicalMeasurement A vol 9 supple-ment pp 15ndash20 1988

[202] M Sperandio M Guermandi and R Guerrieri ldquoA four-shell diffusion phantom of the head for electrical impedancetomographyrdquo IEEE Transactions on Biomedical Engineering vol59 no 2 pp 383ndash389 2012

[203] G Hahn M Beer I Frerichs T Dudykevych T Schroderand G Hellige ldquoA simple method to check the dynamicperformance of electrical impedance tomography systemsrdquoPhysiological Measurement vol 21 no 1 pp 53ndash60 2000

[204] GHahnA Just andGHellige ldquoDetermination of the dynamicmeasurement error of EIT systemsrdquo in Proceedings of the 13thInternational Conference on Electrical Bioimpedance and the 8thConference on Electrical Impedance Tomography (ICEBI rsquo07) pp320ndash323 September 2007

[205] A E Hartinger H Gagnon and R Guardo ldquoA method formodelling and optimizing an electrical impedance tomographysystemrdquo Physiological Measurement vol 27 no 5 pp S51ndashS642006

[206] A E Hartinger H Gagnon and R Guardo ldquoAccounting forhardware imperfections in EIT image reconstruction algo-rithmsrdquo Physiological Measurement vol 28 no 7 pp S13ndashS272007

[207] H Gagnon A E Hartinger A Adler and R Guardo ldquoAresistive mesh phantom for assessing the performance of EITsystemsrdquo in EIT Conference Manchester UK 2009

[208] GHahnA Just J Dittmar andGHellige ldquoSystematic errors ofEIT systems determined by easily-scalable resistive phantomsrdquoPhysiological Measurement vol 29 pp S163ndashS172 2008

[209] H Gagnon Y Sigmen A E Hartinger and R Guardo AnActive Phantom to Assess the Robustness of EIT Systems toElectrode Contact Impedance Variations EIT Manchester UK2009

[210] T K Bera and J Nagaraju ldquoResistivity imaging of a reconfig-urable phantom with circular inhomogeneities in 2D-electricalimpedance tomographyrdquo Measurement vol 44 no 3 pp 518ndash526 2011

[211] T K Bera and J Nagaraju ldquoCommon groundmethod of currentinjection in electrical impedance tomographyrdquo in Proceedings ofthe 4th International Conference on Recent trends in ComputingCommunication amp Information Technologies vol 270 pp 593ndash605 Tamil Nadu India 2011


[212] T K Bera and J Nagaraju ldquoStudying the 2D-image recon-struction of non biological and biological inhomogeneities inElectrical Impedance Tomography (EIT) with EIDORSrdquo inInternational Conference on Advanced Computing Networkingand Security pp 132ndash136 Mangalore India 2011

[213] T K Bera S K Biswas K Rajan and N Jampana ldquoImprovingthe image reconstruction in Electrical Impedance Tomogra-phy (EIT) with block matrix-based Multiple Regularization(BMMR) a practical phantom studyrdquo in Proceedings of theWorld Congress on Information and Communication Technolo-gies (WICT rsquo11) pp 1346ndash1351 Mumbai India December 2011

[214] T K Bera S K Biswas K Rajan and J Nagaraju ldquoImagereconstruction in Electrical Impedance Tomography (EIT)withprojection error propagation-based regularization (PEPR) apractical phantom studyrdquo in Advanced Computing Networkingand Security vol 7135 of Lecture Notes in Computer Science pp95ndash105 2012

[215] T K Bera and J Nagaraju ldquoA multifrequency ElectricalImpedance Tomography (EIT) system for biomedical imagingrdquoin International Conference on Signal Processing and Communi-cations pp 1ndash5 Bangalore India 2012

[216] N K Soni H Dehghani A Hartov and K D Paulsen ldquoA noveldata calibration scheme for electrical impedance tomographyrdquoPhysiological Measurement vol 24 no 2 pp 421ndash435 2003

[217] G QiaoWWang LWang Y He B Bramer andMAl-AkaidildquoInvestigation of biological phantom for 2D and 3D breast EITimagesrdquo in Proceedings of the 13th International Conference onElectrical Bioimpedance and the 8th Conference on ElectricalImpedance Tomography (ICEBI rsquo07) vol 17 part 10 pp 328ndash331Graz Austria September 2007

[218] J Conway ldquoElectrical impedance tomography for thermalmon-itoring of hyperthermia treatment an assessment using in vitroand in vivo measurementsrdquo Clinical Physics and PhysiologicalMeasurement A vol 8 no 4 pp 141ndash146 1987

[219] T K Bera and J Nagaraju ldquoA reconfigurable practical phantomfor studying the 2 D Electrical Impedance Tomography (EIT)using a FEM based forward solverrdquo in Proceedings of the 10thInternational Conference on Biomedical Applications of ElectricalImpedance Tomography School of MathematicsTheUniversityof Manchester Manchester UK June 2009

[220] S Ahn S C Jun J K Seo J Lee E J Woo and D HolderldquoFrequency-difference electrical impedance tomography phan-tom imaging experimentsrdquo Journal of Physics vol 224 no 1Article ID 012152 2010

[221] I D Schneider R Kleffel D Jennings and A J CourtenayldquoDesign of an electrical impedance tomography phantom usingactive elementsrdquo Medical and Biological Engineering and Com-puting vol 38 no 4 pp 390ndash394 2000

[222] L-H Juang and M-N Wu ldquoImage noise reduction usingWiener filteringwith pseudo-inverserdquoMeasurement vol 43 no10 pp 1649ndash1655 2010

[223] M Z Ur Rahman G V S Karthik S Y Fathima and ALay-Ekuakille ldquoAn efficient cardiac signal enhancement usingtime-frequency realization of leaky adaptive noise cancelers forremote health monitoring systemsrdquo Measurement vol 46 no10 pp 3815ndash3835 2013

[224] A Jeanvoine D Gnansia E Truy and C Berger-VachonldquoSource detection influence of the Doerbeckerrsquos noise reduc-tion algorithm Simulation in the case of binaural cochlearimplant codingrdquoMeasurement vol 47 pp 240ndash247 2014

[225] I J Amin A J Taylor F Junejo A Al-Habaibeh and R MParkin ldquoAutomated people-counting by using low-resolution

infrared and visual camerasrdquo Measurement vol 41 no 6 pp589ndash599 2008

[226] Y-H Wang J Qiao J-B Li P Fu S-C Chu and J FRoddick ldquoSparse representation-based MRI super-resolutionreconstructionrdquoMeasurement vol 47 pp 946ndash953 2014

[227] W D Hou and Y L Mo ldquoIncreasing image resolution inelectrical impedance tomographyrdquo Electronics Letters vol 38no 14 pp 701ndash702 2002

[228] E Yoshida H Yamashita H Tashima et al ldquoDesign studyof the DOI-PET scanners with the Xrsquotal cubes toward sub-millimeter spatial resolutionrdquo Journal of Medical Imaging andHealth Informatics vol 3 pp 131ndash134 2013

[229] G Sun C R French G R Martin et al ldquoComparison ofmultifrequency bioelectrical impedance analysis with dual-energyX-ray absorptiometry for assessment of percentage bodyfat in a large healthy populationrdquo American Journal of ClinicalNutrition vol 81 no 1 pp 74ndash78 2005

[230] J R Mager S D Sibley T R Beckman T A Kelloggand C P Earthman ldquoMultifrequency bioelectrical impedanceanalysis and bioimpedance spectroscopy for monitoring fluidand body cellmass changes after gastric bypass surgeryrdquoClinicalNutrition vol 27 no 6 pp 832ndash841 2008

[231] S Gawad T Sun N G Green and H Morgan ldquoImpedancespectroscopy using maximum length sequences application tosingle cell analysisrdquo Review of Scientific Instruments vol 78 no5 Article ID 054301 2007

[232] T Sun D Holmes S Gawad N G Green and H MorganldquoHigh speed multi-frequency impedance analysis of singleparticles in a microfluidic cytometer using maximum lengthsequencesrdquo Lab on a Chip vol 7 no 8 pp 1034ndash1040 2007

[233] T Sun S Gawad C Bernabini N G Green and H Mor-gan ldquoBroadband single cell impedance spectroscopy usingmaximum length sequences theoretical analysis and practicalconsiderationsrdquo Measurement Science and Technology vol 18no 9 pp 2859ndash2868 2007

[234] J M Porter I D Swain and P G Shakespeare ldquoMeasurementof pulsatile limb and finger blood flow by electrical impedanceplethysmography criteria for the diagnosis of abnormal flowrdquoJournal of Biomedical Engineering vol 9 no 4 pp 367ndash373 1987


[236] R Shankar S Y Shao and J G Webster ldquoA Fully Auto-mated Multichannel Digital Electrical Impedance Plethys-mographrdquo 2008 httpcsifaueduwp-contentuploads201301Fully Automated Multi Channel Plethysmographpdf

[237] R Shankar M Gopinathan and J G Webster ldquoDigital SignalProcessing in Clinical Validation Studies with ImpedancePlethysmographyrdquo 2008 httpcsifaueduwp-contentuploads201301Digital Signal Processing Impedance Plethysmogra-phypdf

[238] R Shankar M Martinez and J G Webster Software Proto-col Techniques for Clinical Validation Studies with ImpedancePlethysmography 2008

[239] M Saritha K P Joseph and T M Abraham ldquoAutomaticinterpretation of MRI brain images using probabilistic neuralnetworkrdquo Journal of Medical Imaging and Health Informaticsvol 3 pp 237ndash241 2013

[240] E Schwartz and L Shamir ldquoCorrelation between brainMRI andcontinuous physiological and environmental traits using 2D


global descriptors and multi-order image transformsrdquo Journalof Medical Imaging and Health Informatics vol 3 pp 12ndash162013

[241] E A Zanaty ldquoAn approach based on fusion concepts forimproving brain Magnetic Resonance Images (MRIs) segmen-tationrdquo Journal of Medical Imaging and Health Informatics vol3 pp 30ndash37 2013

[242] Mohanalin P K Kalra and N Kumar ldquoPoissons equationbased image registration an application for matching 2Dmam-mogramsrdquo Journal of Medical Imaging and Health Informaticsvol 4 pp 49ndash57 2014

[243] T M Oliveira T K Brasileiro SantrsquoAnna F M Mauad JElias Jr and V F Muglia ldquoBreast imaging is the sonographicdescriptor of orientation valid for magnetic resonance imag-ingrdquo Journal of Magnetic Resonance Imaging vol 36 no 6 pp1383ndash1388 2012

[244] K K Lindfors J M Boone M S Newell and C J DOrsildquoDedicated breast computed tomography the optimal cross-sectional imaging solutionrdquo Radiologic Clinics of North Amer-ica vol 48 no 5 pp 1043ndash1054 2010

[245] H -K Wu H -S Tseng L -S Chen et al ldquoFlow index of 3Dbreast ultrasound was strongly correlated with axillary lymphnode metastasis in the absence of lymphovascular invasionrdquoJournal of Medical Imaging and Health Informatics vol 4 pp66ndash70 2014

[246] J M Boone A L C Kwan K Yang G W Burkett K KLindfors andT RNelson ldquoComputed tomography for imagingthe breastrdquo Journal of Mammary Gland Biology and Neoplasiavol 11 no 2 pp 103ndash111 2006

[247] E L Rosen W B Eubank and D A Mankoff ldquoFDG PETPETCT and breast cancer imagingrdquo Radiographics vol 27supplement 1 pp S215ndashS229 2007

[248] Y Zhou ldquoUltrasound diagnosis of breast cancerrdquo Journal ofMedical Imaging and Health Informatics vol 3 pp 157ndash1702013

[249] D M Ikeda D R Baker and B L Daniel ldquoMagnetic resonanceimaging of breast cancer clinical indications and breast MRIreporting systemrdquo Journal of Magnetic Resonance Imaging vol12 no 6 pp 975ndash983 2000

[250] A C Silva B G Morse A K Hara R G Paden N Hongoand W Pavlicek ldquoDual-energy (spectral) CT applications inabdominal imagingrdquoRadiographics vol 31 no 4 pp 1031ndash10502011

[251] W A See and L Hoxie ldquoChest staging in testis cancer patientsimaging modality selection based upon risk assessment asdetermined by abdominal computerized tomography scanresultsrdquo Journal of Urology vol 150 no 3 pp 874ndash878 1993

[252] K B Kim H J Park S Kim and S J Lee ldquoExtractionand analysis of muscle areas from abdominal ultrasonographicimagesrdquo Journal of Medical Imaging and Health Informatics vol4 pp 8ndash13 2014

[253] M Gupta D L Schriger J R Hiatt et al ldquoSelective useof computed tomography compared with routine whole bodyimaging in patients with blunt traumardquo Annals of EmergencyMedicine vol 58 no 5 pp 407ndashe15 2011

[254] G P Schmidt H Kramer M F Reiser and C Glaser ldquoWhole-body magnetic resonance imaging and positron emissiontomography-computed tomography in oncologyrdquo Topics inMagnetic Resonance Imaging vol 18 no 3 pp 193ndash202 2007

[255] S J Eustace and E Nelson ldquoWhole body magnetic resonanceimagingrdquo British Medical Journal vol 328 no 7453 pp 1387ndash1388 2004

[256] K U Juergens M L Oei M Weckesser et al ldquoWhole-body imaging of oncologic patients using 16-channel PET-CT evaluation of an IV contrast enhanced MDCT protocolrdquoNuklearMedizin vol 47 no 1 pp 30ndash36 2008

International Journal of

AerospaceEngineeringHindawi Publishing Corporationhttpwwwhindawicom Volume 2014

RoboticsJournal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014


Active and Passive Electronic Components

Control Scienceand Engineering

Journal of



RotatingMachinery


Hindawi Publishing Corporation httpwwwhindawicom

Journal ofEngineeringVolume 2014

Submit your manuscripts athttpwwwhindawicom

VLSI Design



Shock and Vibration


Civil EngineeringAdvances in

Acoustics and VibrationAdvances in



Electrical and Computer Engineering

Journal of

Advances inOptoElectronics


Volume 2014

The Scientific World JournalHindawi Publishing Corporation httpwwwhindawicom Volume 2014

SensorsJournal of


Modelling amp Simulation in EngineeringHindawi Publishing Corporation httpwwwhindawicom Volume 2014


Chemical EngineeringInternational Journal of Antennas and

Propagation




Navigation and Observation



DistributedSensor Networks



provides the impedance variations over frequencies AlsoEIS has been studied for the noninvasive characterizationof biological as well as nonbiological materials in frequencydomain whereas BIA IPG and ICG are used in biologicalfields only On the other hand the BIA IPG and ICG areapplied on biological tissues and generally at a particularfrequency BIA IPG and ICG all are impedance analyzingtechniques which provide the impedance values of the tissuesample as a lumped estimation whereas the bioelectrical EIScalculates and analyzes the electrical impedance at differentfrequencies which enable us to obtain not only the impedancevalues of the tissue sample as a lumped estimation at asuitable frequency (generally 50 kHz) but also it providesthe information to understand several complex bioelectricalphenomena like dielectric relaxation and dielectric disper-sions The information about the dielectric relaxation [40ndash42] and 120572 120573 and 120574 dispersions [40ndash42] enables us to analyzeand understand the complex bioelectrical phenomena [40ndash42] occurring in cells and tissues under an alternating electriccurrent signal BIA EIS IPG and ICG investigate the tissueproperties by assessing the lump impedance parametersobtained from the boundary voltage current measurementOn the other hand EIT provides a spatial distribution (2D or3D) of the impedance profile of a domain under test using aset of boundary voltage-current dataTherefore EIT is foundwith the potential of visualizing the tissue physiology andpathology in terms of tomographic images of the electricalimpedance distribution and hence it has been applied inseveral applications

The paper has discussed about the BIA EIS IPG ICGand EIT techniques and reviewed their technical perspectivealong with their application in different fields The workingprinciple applications merits and demerits of all thesemethods have been discussed in detail in this paper alongwith their other technical issues The paper discusses aboutthe present status challenges and future directions of theimpedance methods

2 Bioelectrical Impedance

21 Biological Tissues and Their Bioelectrical Impedance Theanimals or plants are the living subjects which are developedwith cells and tissues arranged in three-dimensional (3D)space Animals and plants are developed with animal tissuesand plant tissues respectively The animal tissues and theplant tissues are composed of the 3Darrays of animal cells andplant cells respectively Human body is a complex biologicalstructure composed of several tissues [1] composed of the3D arrays of cells All the biological cells such as animalcells and plant cells contain intracellular fluids (ICF) and cellenvelop (cell membrane for animal cells and cell membraneand cell wall for plant cells) and are suspended in extracellularfluids (ECF) The animal cells and plant cells suspended inextracellular fluids (ECF) show a different behavior to analternating electrical signal producing a complex electricalimpedance [2ndash4] which is called bioelectrical impedance orelectrical bioimpedance

The bioelectrical impedance depends on the tissue com-position as well as the frequency of the applied ac signal [2ndash4] and hence the frequency response of the bioelectricalimpedance of the biological tissues not only depends ontheir physiological and physiochemical status but also itvaries with applied signal frequency Moreover the bioelec-trical impedance varies from tissue to tissue and subject tosubject even if it varies for the measurement directions inanisotropic tissue structures Even the complex bioelectricalimpedance changes due to the variation in tissue structurecomposition and health status [5 6] Hence by studying theimpedance analysis of biological tissue one can obtain a lot ofinformation about the tissue anatomy and tissue physiologyTherefore the complex electrical impedance analysis of bio-logical tissues is found to be an efficient tool for noninvasiveinvestigations of physiological or pathological status

22 Origin of the Bioelectrical Impedance The biologicalsubjects are developed with biological tissues which aredeveloped with biological cells arranged in a very complexthree-dimensional arrays suspended in extracellular matrixcalled an extracellular fluids (ECF) [5] In animal tissuethe cells are composed of intracellular fluids (ICF) cellmembranes (CM) which are suspended in extracellular fluids(ECF) whereas the plant cells are composed of ICF CM andcell wall (CW) and are suspended in ECF

ICF CM and ECF all are composed of different materialshaving their distinguished electrical properties and thereforeeach of these cell and tissue components respond distin-guishably to an alternating current signal The intracellularfluids are composed of the cytoplasm and the nucleusThe cytoplasm and the nucleus are mostly made up ofsolution of proteins different chemicals salts and watersand hence these materials are electrically conducting Theextracellular fluids are alsomade up of electrically conductingmaterials As the intracellular fluids and extracellular fluidsin biological tissues are composed of ionic solution and thehighly conducting materials they provide highly conductingpaths (low resistive paths) to the applied current signal [34]But the cell membranes (Figure 1) of the cells in a biologi-cal tissue are composed of electrically nonconducting lipidbilayers (Figure 1(a)) sandwiched between two conductingprotein layers [35 36] and form the protein-lipid-protein(P-L-P) structures In the P-L-P structures (Figure 1(b)) thehydrophobic tails do not absorb the water and hydrophilicheads are attached to the protein layers The P-L-P [35 36]sandwiched structure (Figure 1(c)) provides a capacitance[34] to the applied alternating current signal and contributesto a capacitive reactance

As a result the overall response of the biological tissuesto an alternating electrical signal applied to it producesa complex bioelectrical impedance (119885119887) [2ndash4] which is afunction of tissue composition as well as the frequency ofthe applied ac signal [2ndash4] The impedance 119885119887 is a complexquantity and varies with tissue structure tissue compositiontissue health and signal frequency Therefore the impedancevaries from subject to subject tissue to tissue and sometimewith measurement direction within the same tissue Even


Hyd

roph

obic

(non

pola

r) ta

ils

heads




Hydrophilic zone

Hydrophobic zone

Hydrophilic (polar)

(a)


Phospholipid units


(b)





(c)









Voltmeter

Current source


I

V

(a)

Current source

Voltmeter


I

V

(b)




119885ang (120579) =

119881ang (1205791)

119868ang (1205792)

(1)







Water 64

Protein20

Fat10

Minerals5

Carbohydrate 1


(anatomy model)

(a)

Oxygen 65

Carbon 18

Hydrogen10

Others7


(chemical model)

(b)

















Weight



Men Women


Fat 15ndash20

Minerals 58ndash60

Protein 16ndash18

Fat 20ndash30

Minerals 55ndash60

Protein 14ndash16


(a)

(ICW) asymp 44

(ECW) asymp 29



Visceral proteins

Intracellular water

Extracellular water


water (TBW)


(b)













(a)



Adiposetissues with

less water

(b)



for females

Fat-Free Mass (kg)

= 0475 times [

(ht (cm))2

119877 (ohms)] + 0295 times wt (kg) + 549

(2)

for males

Fat-Free Mass (kg)

= 0485 times [

(ht (cm))2

119877 (ohms)] + 0338 times wt (kg) + 352

(3)





TBW (kg) = 184 + 045(


)

+ 011 (weight)

(5a)


TBW (kg) = 0377(

height2



(5b)


height2



(5c)





BIA

(a)




right foot




hand)


(b)



















AZ =

120588L

A


Z

or Z =120588L2

volumeL



Highfrequency

current

Extracellularfluids





















TBW (Liters) = [


]


(6)

ECW (Liters) = [


]

+ (006895 times weight) + 3794

(7a)

TBW (Liters) = [


]

+ (018782 times weight) + 8197

(7b)





119885 (119891119894) =

119881 (119891119894)

119868 (119891119894)

(8)








Stop


Start




119885119887 = 120588119887

119871

Δ119860

(9)


Δ119881 = 119871 sdot Δ119860 = 120588119887

1198712

119885119887

(10)




Δ119885 = (119885119887 119885) minus 119885 = minus

1198852

119885 + 119885119887

(11)


Current electrodes

Body part

Current source

Voltmeter

Voltageelectrodes

I

V

(a)

Body part

Zb

Z

Z

A

ΔZ

ΔA

L

(b)



1

119885119887

cong minus

Δ119885

1198852 (12)


Δ119881 = minus120588119887

1198712Δ119885

1198852 (13)





















(pulmonary emboli)


Outer electrodes



potentials

Xiphoid

Sternum

Xiphisternal joint

(a)



Vena cavae


Thoracic aorta

TEB and ECGsensing

(b)











R minus Z

Z

10Ωs

Calibration

ECG

PCG

1 2 3 4

Time (s)0

A B XYOZ

dZ

dt

dZ

dtmin









2times 119889119885119889119905 times LVET


Formulae)








120597Ω

Ω

VV

V

VV

I













I

V

V

V

V

V

V

V

V

V

VVV

V

1

2

3456

7

8

9

10

1112 13 14

15

16

(a)

I

V

V

V

V

V

V

V

V

V

VVV

V

1

2

3456

7

8

9

10

1112 13 14

15

16

(b)



























Ω






Mul

tiple

xer m

odul

e

CPU

Phantom withsensors

1

23

456

7

8

9

1011

12 1314

15

16

Y

X

minus025

0

025

minus005 0 005



algorithm

Start

Guess

Stop

CPU

No

Yes


12059001205900

120590kJ(120590k) [Vc]k = Γ(120590k)


Measureddata [Vm]k


ΔV lt 120576

Y

X

minus025

0

025

minus005 0 005








nabla sdot 120590nabla120601 = 0 (14)



Φ = Φ119894 (15)



int

120597Ω

120590

120597Φ

120597119899

=



0 otherwise(16)


[Φ] = [119870 (120590119890)]minus1[119862] (17)




119904119903 =

1

2

1003817100381710038171003817119881119898 minus 119865

1003817100381710038171003817

2+

1

2

120582 119877120590

=

1

2

(119881119898 minus 119865)119879(119881119898 minus 119865) +

1

2

120582(119877120590)119879(119877120590)2

(18)



Δ120590 =

1199041015840

119903

11990410158401015840119903

=

(1198651015840)

119879

(119881119898 minus 119865) minus 120582(119877)119879(119877120590)

(1198651015840)119879(1198651015840) minus (119865

10158401015840)119879(119881119898 minus 119865) + 120582119877

119879119877

(19)


Δ120590 =

119869119879(119881119898 minus 119891) minus 120582119868120590

119869119879119869 + 120582119868

(20)



[120590119896+1] = [120590119896] + [[119869119879119869 + 120582119868]

minus1

[119869119879[119881119898 minus 119891] minus 120582119868120590]]

119896 (21)











9 Conclusions




References

















































































































































































































































































RoboticsJournal of





Journal of



RotatingMachinery





VLSI Design



Shock and Vibration







Journal of



Volume 2014


SensorsJournal of





Propagation










Hyd

roph

obic

(non

pola

r) ta

ils

heads




Hydrophilic zone

Hydrophobic zone

Hydrophilic (polar)

(a)


Phospholipid units


(b)





(c)









Voltmeter

Current source


I

V

(a)

Current source

Voltmeter


I

V

(b)




119885ang (120579) =

119881ang (1205791)

119868ang (1205792)

(1)







Water 64

Protein20

Fat10

Minerals5

Carbohydrate 1


(anatomy model)

(a)

Oxygen 65

Carbon 18

Hydrogen10

Others7


(chemical model)

(b)

















Weight



Men Women


Fat 15ndash20

Minerals 58ndash60

Protein 16ndash18

Fat 20ndash30

Minerals 55ndash60

Protein 14ndash16


(a)

(ICW) asymp 44

(ECW) asymp 29



Visceral proteins

Intracellular water

Extracellular water


water (TBW)


(b)













(a)



Adiposetissues with

less water

(b)



for females

Fat-Free Mass (kg)

= 0475 times [

(ht (cm))2

119877 (ohms)] + 0295 times wt (kg) + 549

(2)

for males

Fat-Free Mass (kg)

= 0485 times [

(ht (cm))2

119877 (ohms)] + 0338 times wt (kg) + 352

(3)





TBW (kg) = 184 + 045(


)

+ 011 (weight)

(5a)


TBW (kg) = 0377(

height2



(5b)


height2



(5c)





BIA

(a)




right foot




hand)


(b)



















AZ =

120588L

A


Z

or Z =120588L2

volumeL



Highfrequency

current

Extracellularfluids





















TBW (Liters) = [


]


(6)

ECW (Liters) = [


]

+ (006895 times weight) + 3794

(7a)

TBW (Liters) = [


]

+ (018782 times weight) + 8197

(7b)





119885 (119891119894) =

119881 (119891119894)

119868 (119891119894)

(8)








Stop


Start




119885119887 = 120588119887

119871

Δ119860

(9)


Δ119881 = 119871 sdot Δ119860 = 120588119887

1198712

119885119887

(10)




Δ119885 = (119885119887 119885) minus 119885 = minus

1198852

119885 + 119885119887

(11)


Current electrodes

Body part

Current source

Voltmeter

Voltageelectrodes

I

V

(a)

Body part

Zb

Z

Z

A

ΔZ

ΔA

L

(b)



1

119885119887

cong minus

Δ119885

1198852 (12)


Δ119881 = minus120588119887

1198712Δ119885

1198852 (13)





















(pulmonary emboli)


Outer electrodes



potentials

Xiphoid

Sternum

Xiphisternal joint

(a)



Vena cavae


Thoracic aorta

TEB and ECGsensing

(b)











R minus Z

Z

10Ωs

Calibration

ECG

PCG

1 2 3 4

Time (s)0

A B XYOZ

dZ

dt

dZ

dtmin









2times 119889119885119889119905 times LVET


Formulae)








120597Ω

Ω

VV

V

VV

I













I

V

V

V

V

V

V

V

V

V

VVV

V

1

2

3456

7

8

9

10

1112 13 14

15

16

(a)

I

V

V

V

V

V

V

V

V

V

VVV

V

1

2

3456

7

8

9

10

1112 13 14

15

16

(b)



























Ω






Mul

tiple

xer m

odul

e

CPU

Phantom withsensors

1

23

456

7

8

9

1011

12 1314

15

16

Y

X

minus025

0

025

minus005 0 005



algorithm

Start

Guess

Stop

CPU

No

Yes


12059001205900

120590kJ(120590k) [Vc]k = Γ(120590k)


Measureddata [Vm]k


ΔV lt 120576

Y

X

minus025

0

025

minus005 0 005








nabla sdot 120590nabla120601 = 0 (14)



Φ = Φ119894 (15)



int

120597Ω

120590

120597Φ

120597119899

=



0 otherwise(16)


[Φ] = [119870 (120590119890)]minus1[119862] (17)




119904119903 =

1

2

1003817100381710038171003817119881119898 minus 119865

1003817100381710038171003817

2+

1

2

120582 119877120590

=

1

2

(119881119898 minus 119865)119879(119881119898 minus 119865) +

1

2

120582(119877120590)119879(119877120590)2

(18)



Δ120590 =

1199041015840

119903

11990410158401015840119903

=

(1198651015840)

119879

(119881119898 minus 119865) minus 120582(119877)119879(119877120590)

(1198651015840)119879(1198651015840) minus (119865

10158401015840)119879(119881119898 minus 119865) + 120582119877

119879119877

(19)


Δ120590 =

119869119879(119881119898 minus 119891) minus 120582119868120590

119869119879119869 + 120582119868

(20)



[120590119896+1] = [120590119896] + [[119869119879119869 + 120582119868]

minus1

[119869119879[119881119898 minus 119891] minus 120582119868120590]]

119896 (21)











9 Conclusions




References

















































































































































































































































































RoboticsJournal of





Journal of



RotatingMachinery





VLSI Design



Shock and Vibration







Journal of



Volume 2014


SensorsJournal of





Propagation










Voltmeter

Current source


I

V

(a)

Current source

Voltmeter


I

V

(b)




119885ang (120579) =

119881ang (1205791)

119868ang (1205792)

(1)







Water 64

Protein20

Fat10

Minerals5

Carbohydrate 1


(anatomy model)

(a)

Oxygen 65

Carbon 18

Hydrogen10

Others7


(chemical model)

(b)

















Weight



Men Women


Fat 15ndash20

Minerals 58ndash60

Protein 16ndash18

Fat 20ndash30

Minerals 55ndash60

Protein 14ndash16


(a)

(ICW) asymp 44

(ECW) asymp 29



Visceral proteins

Intracellular water

Extracellular water


water (TBW)


(b)













(a)



Adiposetissues with

less water

(b)



for females

Fat-Free Mass (kg)

= 0475 times [

(ht (cm))2

119877 (ohms)] + 0295 times wt (kg) + 549

(2)

for males

Fat-Free Mass (kg)

= 0485 times [

(ht (cm))2

119877 (ohms)] + 0338 times wt (kg) + 352

(3)





TBW (kg) = 184 + 045(


)

+ 011 (weight)

(5a)


TBW (kg) = 0377(

height2



(5b)


height2



(5c)





BIA

(a)




right foot




hand)


(b)



















AZ =

120588L

A


Z

or Z =120588L2

volumeL



Highfrequency

current

Extracellularfluids





















TBW (Liters) = [


]


(6)

ECW (Liters) = [


]

+ (006895 times weight) + 3794

(7a)

TBW (Liters) = [


]

+ (018782 times weight) + 8197

(7b)





119885 (119891119894) =

119881 (119891119894)

119868 (119891119894)

(8)








Stop


Start




119885119887 = 120588119887

119871

Δ119860

(9)


Δ119881 = 119871 sdot Δ119860 = 120588119887

1198712

119885119887

(10)




Δ119885 = (119885119887 119885) minus 119885 = minus

1198852

119885 + 119885119887

(11)


Current electrodes

Body part

Current source

Voltmeter

Voltageelectrodes

I

V

(a)

Body part

Zb

Z

Z

A

ΔZ

ΔA

L

(b)



1

119885119887

cong minus

Δ119885

1198852 (12)


Δ119881 = minus120588119887

1198712Δ119885

1198852 (13)





















(pulmonary emboli)


Outer electrodes



potentials

Xiphoid

Sternum

Xiphisternal joint

(a)



Vena cavae


Thoracic aorta

TEB and ECGsensing

(b)











R minus Z

Z

10Ωs

Calibration

ECG

PCG

1 2 3 4

Time (s)0

A B XYOZ

dZ

dt

dZ

dtmin









2times 119889119885119889119905 times LVET


Formulae)








120597Ω

Ω

VV

V

VV

I













I

V

V

V

V

V

V

V

V

V

VVV

V

1

2

3456

7

8

9

10

1112 13 14

15

16

(a)

I

V

V

V

V

V

V

V

V

V

VVV

V

1

2

3456

7

8

9

10

1112 13 14

15

16

(b)



























Ω






Mul

tiple

xer m

odul

e

CPU

Phantom withsensors

1

23

456

7

8

9

1011

12 1314

15

16

Y

X

minus025

0

025

minus005 0 005



algorithm

Start

Guess

Stop

CPU

No

Yes


12059001205900

120590kJ(120590k) [Vc]k = Γ(120590k)


Measureddata [Vm]k


ΔV lt 120576

Y

X

minus025

0

025

minus005 0 005








nabla sdot 120590nabla120601 = 0 (14)



Φ = Φ119894 (15)



int

120597Ω

120590

120597Φ

120597119899

=



0 otherwise(16)


[Φ] = [119870 (120590119890)]minus1[119862] (17)




119904119903 =

1

2

1003817100381710038171003817119881119898 minus 119865

1003817100381710038171003817

2+

1

2

120582 119877120590

=

1

2

(119881119898 minus 119865)119879(119881119898 minus 119865) +

1

2

120582(119877120590)119879(119877120590)2

(18)



Δ120590 =

1199041015840

119903

11990410158401015840119903

=

(1198651015840)

119879

(119881119898 minus 119865) minus 120582(119877)119879(119877120590)

(1198651015840)119879(1198651015840) minus (119865

10158401015840)119879(119881119898 minus 119865) + 120582119877

119879119877

(19)


Δ120590 =

119869119879(119881119898 minus 119891) minus 120582119868120590

119869119879119869 + 120582119868

(20)



[120590119896+1] = [120590119896] + [[119869119879119869 + 120582119868]

minus1

[119869119879[119881119898 minus 119891] minus 120582119868120590]]

119896 (21)











9 Conclusions




References

















































































































































































































































































RoboticsJournal of





Journal of



RotatingMachinery





VLSI Design



Shock and Vibration







Journal of



Volume 2014


SensorsJournal of





Propagation










Water 64

Protein20

Fat10

Minerals5

Carbohydrate 1


(anatomy model)

(a)

Oxygen 65

Carbon 18

Hydrogen10

Others7


(chemical model)

(b)

















Weight



Men Women


Fat 15ndash20

Minerals 58ndash60

Protein 16ndash18

Fat 20ndash30

Minerals 55ndash60

Protein 14ndash16


(a)

(ICW) asymp 44

(ECW) asymp 29



Visceral proteins

Intracellular water

Extracellular water


water (TBW)


(b)













(a)



Adiposetissues with

less water

(b)



for females

Fat-Free Mass (kg)

= 0475 times [

(ht (cm))2

119877 (ohms)] + 0295 times wt (kg) + 549

(2)

for males

Fat-Free Mass (kg)

= 0485 times [

(ht (cm))2

119877 (ohms)] + 0338 times wt (kg) + 352

(3)





TBW (kg) = 184 + 045(


)

+ 011 (weight)

(5a)


TBW (kg) = 0377(

height2



(5b)


height2



(5c)





BIA

(a)




right foot




hand)


(b)



















AZ =

120588L

A


Z

or Z =120588L2

volumeL



Highfrequency

current

Extracellularfluids





















TBW (Liters) = [


]


(6)

ECW (Liters) = [


]

+ (006895 times weight) + 3794

(7a)

TBW (Liters) = [


]

+ (018782 times weight) + 8197

(7b)





119885 (119891119894) =

119881 (119891119894)

119868 (119891119894)

(8)








Stop


Start




119885119887 = 120588119887

119871

Δ119860

(9)


Δ119881 = 119871 sdot Δ119860 = 120588119887

1198712

119885119887

(10)




Δ119885 = (119885119887 119885) minus 119885 = minus

1198852

119885 + 119885119887

(11)


Current electrodes

Body part

Current source

Voltmeter

Voltageelectrodes

I

V

(a)

Body part

Zb

Z

Z

A

ΔZ

ΔA

L

(b)



1

119885119887

cong minus

Δ119885

1198852 (12)


Δ119881 = minus120588119887

1198712Δ119885

1198852 (13)





















(pulmonary emboli)


Outer electrodes



potentials

Xiphoid

Sternum

Xiphisternal joint

(a)



Vena cavae


Thoracic aorta

TEB and ECGsensing

(b)











R minus Z

Z

10Ωs

Calibration

ECG

PCG

1 2 3 4

Time (s)0

A B XYOZ

dZ

dt

dZ

dtmin









2times 119889119885119889119905 times LVET


Formulae)








120597Ω

Ω

VV

V

VV

I













I

V

V

V

V

V

V

V

V

V

VVV

V

1

2

3456

7

8

9

10

1112 13 14

15

16

(a)

I

V

V

V

V

V

V

V

V

V

VVV

V

1

2

3456

7

8

9

10

1112 13 14

15

16

(b)



























Ω






Mul

tiple

xer m

odul

e

CPU

Phantom withsensors

1

23

456

7

8

9

1011

12 1314

15

16

Y

X

minus025

0

025

minus005 0 005



algorithm

Start

Guess

Stop

CPU

No

Yes


12059001205900

120590kJ(120590k) [Vc]k = Γ(120590k)


Measureddata [Vm]k


ΔV lt 120576

Y

X

minus025

0

025

minus005 0 005








nabla sdot 120590nabla120601 = 0 (14)



Φ = Φ119894 (15)



int

120597Ω

120590

120597Φ

120597119899

=



0 otherwise(16)


[Φ] = [119870 (120590119890)]minus1[119862] (17)




119904119903 =

1

2

1003817100381710038171003817119881119898 minus 119865

1003817100381710038171003817

2+

1

2

120582 119877120590

=

1

2

(119881119898 minus 119865)119879(119881119898 minus 119865) +

1

2

120582(119877120590)119879(119877120590)2

(18)



Δ120590 =

1199041015840

119903

11990410158401015840119903

=

(1198651015840)

119879

(119881119898 minus 119865) minus 120582(119877)119879(119877120590)

(1198651015840)119879(1198651015840) minus (119865

10158401015840)119879(119881119898 minus 119865) + 120582119877

119879119877

(19)


Δ120590 =

119869119879(119881119898 minus 119891) minus 120582119868120590

119869119879119869 + 120582119868

(20)



[120590119896+1] = [120590119896] + [[119869119879119869 + 120582119868]

minus1

[119869119879[119881119898 minus 119891] minus 120582119868120590]]

119896 (21)











9 Conclusions




References

















































































































































































































































































RoboticsJournal of





Journal of



RotatingMachinery





VLSI Design



Shock and Vibration







Journal of



Volume 2014


SensorsJournal of





Propagation










Men Women


Fat 15ndash20

Minerals 58ndash60

Protein 16ndash18

Fat 20ndash30

Minerals 55ndash60

Protein 14ndash16


(a)

(ICW) asymp 44

(ECW) asymp 29



Visceral proteins

Intracellular water

Extracellular water


water (TBW)


(b)













(a)



Adiposetissues with

less water

(b)



for females

Fat-Free Mass (kg)

= 0475 times [

(ht (cm))2

119877 (ohms)] + 0295 times wt (kg) + 549

(2)

for males

Fat-Free Mass (kg)

= 0485 times [

(ht (cm))2

119877 (ohms)] + 0338 times wt (kg) + 352

(3)





TBW (kg) = 184 + 045(


)

+ 011 (weight)

(5a)


TBW (kg) = 0377(

height2



(5b)


height2



(5c)





BIA

(a)




right foot




hand)


(b)



















AZ =

120588L

A


Z

or Z =120588L2

volumeL



Highfrequency

current

Extracellularfluids





















TBW (Liters) = [


]


(6)

ECW (Liters) = [


]

+ (006895 times weight) + 3794

(7a)

TBW (Liters) = [


]

+ (018782 times weight) + 8197

(7b)





119885 (119891119894) =

119881 (119891119894)

119868 (119891119894)

(8)








Stop


Start




119885119887 = 120588119887

119871

Δ119860

(9)


Δ119881 = 119871 sdot Δ119860 = 120588119887

1198712

119885119887

(10)




Δ119885 = (119885119887 119885) minus 119885 = minus

1198852

119885 + 119885119887

(11)


Current electrodes

Body part

Current source

Voltmeter

Voltageelectrodes

I

V

(a)

Body part

Zb

Z

Z

A

ΔZ

ΔA

L

(b)



1

119885119887

cong minus

Δ119885

1198852 (12)


Δ119881 = minus120588119887

1198712Δ119885

1198852 (13)





















(pulmonary emboli)


Outer electrodes



potentials

Xiphoid

Sternum

Xiphisternal joint

(a)



Vena cavae


Thoracic aorta

TEB and ECGsensing

(b)











R minus Z

Z

10Ωs

Calibration

ECG

PCG

1 2 3 4

Time (s)0

A B XYOZ

dZ

dt

dZ

dtmin









2times 119889119885119889119905 times LVET


Formulae)








120597Ω

Ω

VV

V

VV

I













I

V

V

V

V

V

V

V

V

V

VVV

V

1

2

3456

7

8

9

10

1112 13 14

15

16

(a)

I

V

V

V

V

V

V

V

V

V

VVV

V

1

2

3456

7

8

9

10

1112 13 14

15

16

(b)



























Ω






Mul

tiple

xer m

odul

e

CPU

Phantom withsensors

1

23

456

7

8

9

1011

12 1314

15

16

Y

X

minus025

0

025

minus005 0 005



algorithm

Start

Guess

Stop

CPU

No

Yes


12059001205900

120590kJ(120590k) [Vc]k = Γ(120590k)


Measureddata [Vm]k


ΔV lt 120576

Y

X

minus025

0

025

minus005 0 005








nabla sdot 120590nabla120601 = 0 (14)



Φ = Φ119894 (15)



int

120597Ω

120590

120597Φ

120597119899

=



0 otherwise(16)


[Φ] = [119870 (120590119890)]minus1[119862] (17)




119904119903 =

1

2

1003817100381710038171003817119881119898 minus 119865

1003817100381710038171003817

2+

1

2

120582 119877120590

=

1

2

(119881119898 minus 119865)119879(119881119898 minus 119865) +

1

2

120582(119877120590)119879(119877120590)2

(18)



Δ120590 =

1199041015840

119903

11990410158401015840119903

=

(1198651015840)

119879

(119881119898 minus 119865) minus 120582(119877)119879(119877120590)

(1198651015840)119879(1198651015840) minus (119865

10158401015840)119879(119881119898 minus 119865) + 120582119877

119879119877

(19)


Δ120590 =

119869119879(119881119898 minus 119891) minus 120582119868120590

119869119879119869 + 120582119868

(20)



[120590119896+1] = [120590119896] + [[119869119879119869 + 120582119868]

minus1

[119869119879[119881119898 minus 119891] minus 120582119868120590]]

119896 (21)











9 Conclusions




References

















































































































































































































































































RoboticsJournal of





Journal of



RotatingMachinery





VLSI Design



Shock and Vibration







Journal of



Volume 2014


SensorsJournal of





Propagation













(a)



Adiposetissues with

less water

(b)



for females

Fat-Free Mass (kg)

= 0475 times [

(ht (cm))2

119877 (ohms)] + 0295 times wt (kg) + 549

(2)

for males

Fat-Free Mass (kg)

= 0485 times [

(ht (cm))2

119877 (ohms)] + 0338 times wt (kg) + 352

(3)





TBW (kg) = 184 + 045(


)

+ 011 (weight)

(5a)


TBW (kg) = 0377(

height2



(5b)


height2



(5c)





BIA

(a)




right foot




hand)


(b)



















AZ =

120588L

A


Z

or Z =120588L2

volumeL



Highfrequency

current

Extracellularfluids





















TBW (Liters) = [


]


(6)

ECW (Liters) = [


]

+ (006895 times weight) + 3794

(7a)

TBW (Liters) = [


]

+ (018782 times weight) + 8197

(7b)





119885 (119891119894) =

119881 (119891119894)

119868 (119891119894)

(8)








Stop


Start




119885119887 = 120588119887

119871

Δ119860

(9)


Δ119881 = 119871 sdot Δ119860 = 120588119887

1198712

119885119887

(10)




Δ119885 = (119885119887 119885) minus 119885 = minus

1198852

119885 + 119885119887

(11)


Current electrodes

Body part

Current source

Voltmeter

Voltageelectrodes

I

V

(a)

Body part

Zb

Z

Z

A

ΔZ

ΔA

L

(b)



1

119885119887

cong minus

Δ119885

1198852 (12)


Δ119881 = minus120588119887

1198712Δ119885

1198852 (13)





















(pulmonary emboli)


Outer electrodes



potentials

Xiphoid

Sternum

Xiphisternal joint

(a)



Vena cavae


Thoracic aorta

TEB and ECGsensing

(b)











R minus Z

Z

10Ωs

Calibration

ECG

PCG

1 2 3 4

Time (s)0

A B XYOZ

dZ

dt

dZ

dtmin









2times 119889119885119889119905 times LVET


Formulae)








120597Ω

Ω

VV

V

VV

I













I

V

V

V

V

V

V

V

V

V

VVV

V

1

2

3456

7

8

9

10

1112 13 14

15

16

(a)

I

V

V

V

V

V

V

V

V

V

VVV

V

1

2

3456

7

8

9

10

1112 13 14

15

16

(b)



























Ω






Mul

tiple

xer m

odul

e

CPU

Phantom withsensors

1

23

456

7

8

9

1011

12 1314

15

16

Y

X

minus025

0

025

minus005 0 005



algorithm

Start

Guess

Stop

CPU

No

Yes


12059001205900

120590kJ(120590k) [Vc]k = Γ(120590k)


Measureddata [Vm]k


ΔV lt 120576

Y

X

minus025

0

025

minus005 0 005








nabla sdot 120590nabla120601 = 0 (14)



Φ = Φ119894 (15)



int

120597Ω

120590

120597Φ

120597119899

=



0 otherwise(16)


[Φ] = [119870 (120590119890)]minus1[119862] (17)




119904119903 =

1

2

1003817100381710038171003817119881119898 minus 119865

1003817100381710038171003817

2+

1

2

120582 119877120590

=

1

2

(119881119898 minus 119865)119879(119881119898 minus 119865) +

1

2

120582(119877120590)119879(119877120590)2

(18)



Δ120590 =

1199041015840

119903

11990410158401015840119903

=

(1198651015840)

119879

(119881119898 minus 119865) minus 120582(119877)119879(119877120590)

(1198651015840)119879(1198651015840) minus (119865

10158401015840)119879(119881119898 minus 119865) + 120582119877

119879119877

(19)


Δ120590 =

119869119879(119881119898 minus 119891) minus 120582119868120590

119869119879119869 + 120582119868

(20)



[120590119896+1] = [120590119896] + [[119869119879119869 + 120582119868]

minus1

[119869119879[119881119898 minus 119891] minus 120582119868120590]]

119896 (21)











9 Conclusions




References

















































































































































































































































































RoboticsJournal of





Journal of



RotatingMachinery





VLSI Design



Shock and Vibration







Journal of



Volume 2014


SensorsJournal of





Propagation










BIA

(a)




right foot




hand)


(b)



















AZ =

120588L

A


Z

or Z =120588L2

volumeL



Highfrequency

current

Extracellularfluids





















TBW (Liters) = [


]


(6)

ECW (Liters) = [


]

+ (006895 times weight) + 3794

(7a)

TBW (Liters) = [


]

+ (018782 times weight) + 8197

(7b)





119885 (119891119894) =

119881 (119891119894)

119868 (119891119894)

(8)








Stop


Start




119885119887 = 120588119887

119871

Δ119860

(9)


Δ119881 = 119871 sdot Δ119860 = 120588119887

1198712

119885119887

(10)




Δ119885 = (119885119887 119885) minus 119885 = minus

1198852

119885 + 119885119887

(11)


Current electrodes

Body part

Current source

Voltmeter

Voltageelectrodes

I

V

(a)

Body part

Zb

Z

Z

A

ΔZ

ΔA

L

(b)



1

119885119887

cong minus

Δ119885

1198852 (12)


Δ119881 = minus120588119887

1198712Δ119885

1198852 (13)





















(pulmonary emboli)


Outer electrodes



potentials

Xiphoid

Sternum

Xiphisternal joint

(a)



Vena cavae


Thoracic aorta

TEB and ECGsensing

(b)











R minus Z

Z

10Ωs

Calibration

ECG

PCG

1 2 3 4

Time (s)0

A B XYOZ

dZ

dt

dZ

dtmin









2times 119889119885119889119905 times LVET


Formulae)








120597Ω

Ω

VV

V

VV

I













I

V

V

V

V

V

V

V

V

V

VVV

V

1

2

3456

7

8

9

10

1112 13 14

15

16

(a)

I

V

V

V

V

V

V

V

V

V

VVV

V

1

2

3456

7

8

9

10

1112 13 14

15

16

(b)



























Ω






Mul

tiple

xer m

odul

e

CPU

Phantom withsensors

1

23

456

7

8

9

1011

12 1314

15

16

Y

X

minus025

0

025

minus005 0 005



algorithm

Start

Guess

Stop

CPU

No

Yes


12059001205900

120590kJ(120590k) [Vc]k = Γ(120590k)


Measureddata [Vm]k


ΔV lt 120576

Y

X

minus025

0

025

minus005 0 005








nabla sdot 120590nabla120601 = 0 (14)



Φ = Φ119894 (15)



int

120597Ω

120590

120597Φ

120597119899

=



0 otherwise(16)


[Φ] = [119870 (120590119890)]minus1[119862] (17)




119904119903 =

1

2

1003817100381710038171003817119881119898 minus 119865

1003817100381710038171003817

2+

1

2

120582 119877120590

=

1

2

(119881119898 minus 119865)119879(119881119898 minus 119865) +

1

2

120582(119877120590)119879(119877120590)2

(18)



Δ120590 =

1199041015840

119903

11990410158401015840119903

=

(1198651015840)

119879

(119881119898 minus 119865) minus 120582(119877)119879(119877120590)

(1198651015840)119879(1198651015840) minus (119865

10158401015840)119879(119881119898 minus 119865) + 120582119877

119879119877

(19)


Δ120590 =

119869119879(119881119898 minus 119891) minus 120582119868120590

119869119879119869 + 120582119868

(20)



[120590119896+1] = [120590119896] + [[119869119879119869 + 120582119868]

minus1

[119869119879[119881119898 minus 119891] minus 120582119868120590]]

119896 (21)











9 Conclusions




References

















































































































































































































































































RoboticsJournal of





Journal of



RotatingMachinery





VLSI Design



Shock and Vibration







Journal of



Volume 2014


SensorsJournal of





Propagation










AZ =

120588L

A


Z

or Z =120588L2

volumeL



Highfrequency

current

Extracellularfluids





















TBW (Liters) = [


]


(6)

ECW (Liters) = [


]

+ (006895 times weight) + 3794

(7a)

TBW (Liters) = [


]

+ (018782 times weight) + 8197

(7b)





119885 (119891119894) =

119881 (119891119894)

119868 (119891119894)

(8)








Stop


Start




119885119887 = 120588119887

119871

Δ119860

(9)


Δ119881 = 119871 sdot Δ119860 = 120588119887

1198712

119885119887

(10)




Δ119885 = (119885119887 119885) minus 119885 = minus

1198852

119885 + 119885119887

(11)


Current electrodes

Body part

Current source

Voltmeter

Voltageelectrodes

I

V

(a)

Body part

Zb

Z

Z

A

ΔZ

ΔA

L

(b)



1

119885119887

cong minus

Δ119885

1198852 (12)


Δ119881 = minus120588119887

1198712Δ119885

1198852 (13)





















(pulmonary emboli)


Outer electrodes



potentials

Xiphoid

Sternum

Xiphisternal joint

(a)



Vena cavae


Thoracic aorta

TEB and ECGsensing

(b)











R minus Z

Z

10Ωs

Calibration

ECG

PCG

1 2 3 4

Time (s)0

A B XYOZ

dZ

dt

dZ

dtmin









2times 119889119885119889119905 times LVET


Formulae)








120597Ω

Ω

VV

V

VV

I













I

V

V

V

V

V

V

V

V

V

VVV

V

1

2

3456

7

8

9

10

1112 13 14

15

16

(a)

I

V

V

V

V

V

V

V

V

V

VVV

V

1

2

3456

7

8

9

10

1112 13 14

15

16

(b)



























Ω






Mul

tiple

xer m

odul

e

CPU

Phantom withsensors

1

23

456

7

8

9

1011

12 1314

15

16

Y

X

minus025

0

025

minus005 0 005



algorithm

Start

Guess

Stop

CPU

No

Yes


12059001205900

120590kJ(120590k) [Vc]k = Γ(120590k)


Measureddata [Vm]k


ΔV lt 120576

Y

X

minus025

0

025

minus005 0 005








nabla sdot 120590nabla120601 = 0 (14)



Φ = Φ119894 (15)



int

120597Ω

120590

120597Φ

120597119899

=



0 otherwise(16)


[Φ] = [119870 (120590119890)]minus1[119862] (17)




119904119903 =

1

2

1003817100381710038171003817119881119898 minus 119865

1003817100381710038171003817

2+

1

2

120582 119877120590

=

1

2

(119881119898 minus 119865)119879(119881119898 minus 119865) +

1

2

120582(119877120590)119879(119877120590)2

(18)



Δ120590 =

1199041015840

119903

11990410158401015840119903

=

(1198651015840)

119879

(119881119898 minus 119865) minus 120582(119877)119879(119877120590)

(1198651015840)119879(1198651015840) minus (119865

10158401015840)119879(119881119898 minus 119865) + 120582119877

119879119877

(19)


Δ120590 =

119869119879(119881119898 minus 119891) minus 120582119868120590

119869119879119869 + 120582119868

(20)



[120590119896+1] = [120590119896] + [[119869119879119869 + 120582119868]

minus1

[119869119879[119881119898 minus 119891] minus 120582119868120590]]

119896 (21)











9 Conclusions




References

















































































































































































































































































RoboticsJournal of





Journal of



RotatingMachinery





VLSI Design



Shock and Vibration







Journal of



Volume 2014


SensorsJournal of





Propagation











119885 (119891119894) =

119881 (119891119894)

119868 (119891119894)

(8)








Stop


Start




119885119887 = 120588119887

119871

Δ119860

(9)


Δ119881 = 119871 sdot Δ119860 = 120588119887

1198712

119885119887

(10)




Δ119885 = (119885119887 119885) minus 119885 = minus

1198852

119885 + 119885119887

(11)


Current electrodes

Body part

Current source

Voltmeter

Voltageelectrodes

I

V

(a)

Body part

Zb

Z

Z

A

ΔZ

ΔA

L

(b)



1

119885119887

cong minus

Δ119885

1198852 (12)


Δ119881 = minus120588119887

1198712Δ119885

1198852 (13)





















(pulmonary emboli)


Outer electrodes



potentials

Xiphoid

Sternum

Xiphisternal joint

(a)



Vena cavae


Thoracic aorta

TEB and ECGsensing

(b)











R minus Z

Z

10Ωs

Calibration

ECG

PCG

1 2 3 4

Time (s)0

A B XYOZ

dZ

dt

dZ

dtmin









2times 119889119885119889119905 times LVET


Formulae)








120597Ω

Ω

VV

V

VV

I













I

V

V

V

V

V

V

V

V

V

VVV

V

1

2

3456

7

8

9

10

1112 13 14

15

16

(a)

I

V

V

V

V

V

V

V

V

V

VVV

V

1

2

3456

7

8

9

10

1112 13 14

15

16

(b)



























Ω






Mul

tiple

xer m

odul

e

CPU

Phantom withsensors

1

23

456

7

8

9

1011

12 1314

15

16

Y

X

minus025

0

025

minus005 0 005



algorithm

Start

Guess

Stop

CPU

No

Yes


12059001205900

120590kJ(120590k) [Vc]k = Γ(120590k)


Measureddata [Vm]k


ΔV lt 120576

Y

X

minus025

0

025

minus005 0 005








nabla sdot 120590nabla120601 = 0 (14)



Φ = Φ119894 (15)



int

120597Ω

120590

120597Φ

120597119899

=



0 otherwise(16)


[Φ] = [119870 (120590119890)]minus1[119862] (17)




119904119903 =

1

2

1003817100381710038171003817119881119898 minus 119865

1003817100381710038171003817

2+

1

2

120582 119877120590

=

1

2

(119881119898 minus 119865)119879(119881119898 minus 119865) +

1

2

120582(119877120590)119879(119877120590)2

(18)



Δ120590 =

1199041015840

119903

11990410158401015840119903

=

(1198651015840)

119879

(119881119898 minus 119865) minus 120582(119877)119879(119877120590)

(1198651015840)119879(1198651015840) minus (119865

10158401015840)119879(119881119898 minus 119865) + 120582119877

119879119877

(19)


Δ120590 =

119869119879(119881119898 minus 119891) minus 120582119868120590

119869119879119869 + 120582119868

(20)



[120590119896+1] = [120590119896] + [[119869119879119869 + 120582119868]

minus1

[119869119879[119881119898 minus 119891] minus 120582119868120590]]

119896 (21)











9 Conclusions




References

















































































































































































































































































RoboticsJournal of





Journal of



RotatingMachinery





VLSI Design



Shock and Vibration







Journal of



Volume 2014


SensorsJournal of





Propagation










Current electrodes

Body part

Current source

Voltmeter

Voltageelectrodes

I

V

(a)

Body part

Zb

Z

Z

A

ΔZ

ΔA

L

(b)



1

119885119887

cong minus

Δ119885

1198852 (12)


Δ119881 = minus120588119887

1198712Δ119885

1198852 (13)





















(pulmonary emboli)


Outer electrodes



potentials

Xiphoid

Sternum

Xiphisternal joint

(a)



Vena cavae


Thoracic aorta

TEB and ECGsensing

(b)











R minus Z

Z

10Ωs

Calibration

ECG

PCG

1 2 3 4

Time (s)0

A B XYOZ

dZ

dt

dZ

dtmin









2times 119889119885119889119905 times LVET


Formulae)








120597Ω

Ω

VV

V

VV

I













I

V

V

V

V

V

V

V

V

V

VVV

V

1

2

3456

7

8

9

10

1112 13 14

15

16

(a)

I

V

V

V

V

V

V

V

V

V

VVV

V

1

2

3456

7

8

9

10

1112 13 14

15

16

(b)



























Ω






Mul

tiple

xer m

odul

e

CPU

Phantom withsensors

1

23

456

7

8

9

1011

12 1314

15

16

Y

X

minus025

0

025

minus005 0 005



algorithm

Start

Guess

Stop

CPU

No

Yes


12059001205900

120590kJ(120590k) [Vc]k = Γ(120590k)


Measureddata [Vm]k


ΔV lt 120576

Y

X

minus025

0

025

minus005 0 005








nabla sdot 120590nabla120601 = 0 (14)



Φ = Φ119894 (15)



int

120597Ω

120590

120597Φ

120597119899

=



0 otherwise(16)


[Φ] = [119870 (120590119890)]minus1[119862] (17)




119904119903 =

1

2

1003817100381710038171003817119881119898 minus 119865

1003817100381710038171003817

2+

1

2

120582 119877120590

=

1

2

(119881119898 minus 119865)119879(119881119898 minus 119865) +

1

2

120582(119877120590)119879(119877120590)2

(18)



Δ120590 =

1199041015840

119903

11990410158401015840119903

=

(1198651015840)

119879

(119881119898 minus 119865) minus 120582(119877)119879(119877120590)

(1198651015840)119879(1198651015840) minus (119865

10158401015840)119879(119881119898 minus 119865) + 120582119877

119879119877

(19)


Δ120590 =

119869119879(119881119898 minus 119891) minus 120582119868120590

119869119879119869 + 120582119868

(20)



[120590119896+1] = [120590119896] + [[119869119879119869 + 120582119868]

minus1

[119869119879[119881119898 minus 119891] minus 120582119868120590]]

119896 (21)











9 Conclusions




References

















































































































































































































































































RoboticsJournal of





Journal of



RotatingMachinery





VLSI Design



Shock and Vibration







Journal of



Volume 2014


SensorsJournal of





Propagation










Outer electrodes



potentials

Xiphoid

Sternum

Xiphisternal joint

(a)



Vena cavae


Thoracic aorta

TEB and ECGsensing

(b)











R minus Z

Z

10Ωs

Calibration

ECG

PCG

1 2 3 4

Time (s)0

A B XYOZ

dZ

dt

dZ

dtmin









2times 119889119885119889119905 times LVET


Formulae)








120597Ω

Ω

VV

V

VV

I













I

V

V

V

V

V

V

V

V

V

VVV

V

1

2

3456

7

8

9

10

1112 13 14

15

16

(a)

I

V

V

V

V

V

V

V

V

V

VVV

V

1

2

3456

7

8

9

10

1112 13 14

15

16

(b)



























Ω






Mul

tiple

xer m

odul

e

CPU

Phantom withsensors

1

23

456

7

8

9

1011

12 1314

15

16

Y

X

minus025

0

025

minus005 0 005



algorithm

Start

Guess

Stop

CPU

No

Yes


12059001205900

120590kJ(120590k) [Vc]k = Γ(120590k)


Measureddata [Vm]k


ΔV lt 120576

Y

X

minus025

0

025

minus005 0 005








nabla sdot 120590nabla120601 = 0 (14)



Φ = Φ119894 (15)



int

120597Ω

120590

120597Φ

120597119899

=



0 otherwise(16)


[Φ] = [119870 (120590119890)]minus1[119862] (17)




119904119903 =

1

2

1003817100381710038171003817119881119898 minus 119865

1003817100381710038171003817

2+

1

2

120582 119877120590

=

1

2

(119881119898 minus 119865)119879(119881119898 minus 119865) +

1

2

120582(119877120590)119879(119877120590)2

(18)



Δ120590 =

1199041015840

119903

11990410158401015840119903

=

(1198651015840)

119879

(119881119898 minus 119865) minus 120582(119877)119879(119877120590)

(1198651015840)119879(1198651015840) minus (119865

10158401015840)119879(119881119898 minus 119865) + 120582119877

119879119877

(19)


Δ120590 =

119869119879(119881119898 minus 119891) minus 120582119868120590

119869119879119869 + 120582119868

(20)



[120590119896+1] = [120590119896] + [[119869119879119869 + 120582119868]

minus1

[119869119879[119881119898 minus 119891] minus 120582119868120590]]

119896 (21)











9 Conclusions




References

















































































































































































































































































RoboticsJournal of





Journal of



RotatingMachinery





VLSI Design



Shock and Vibration







Journal of



Volume 2014


SensorsJournal of





Propagation










R minus Z

Z

10Ωs

Calibration

ECG

PCG

1 2 3 4

Time (s)0

A B XYOZ

dZ

dt

dZ

dtmin









2times 119889119885119889119905 times LVET


Formulae)








120597Ω

Ω

VV

V

VV

I













I

V

V

V

V

V

V

V

V

V

VVV

V

1

2

3456

7

8

9

10

1112 13 14

15

16

(a)

I

V

V

V

V

V

V

V

V

V

VVV

V

1

2

3456

7

8

9

10

1112 13 14

15

16

(b)



























Ω






Mul

tiple

xer m

odul

e

CPU

Phantom withsensors

1

23

456

7

8

9

1011

12 1314

15

16

Y

X

minus025

0

025

minus005 0 005



algorithm

Start

Guess

Stop

CPU

No

Yes


12059001205900

120590kJ(120590k) [Vc]k = Γ(120590k)


Measureddata [Vm]k


ΔV lt 120576

Y

X

minus025

0

025

minus005 0 005








nabla sdot 120590nabla120601 = 0 (14)



Φ = Φ119894 (15)



int

120597Ω

120590

120597Φ

120597119899

=



0 otherwise(16)


[Φ] = [119870 (120590119890)]minus1[119862] (17)




119904119903 =

1

2

1003817100381710038171003817119881119898 minus 119865

1003817100381710038171003817

2+

1

2

120582 119877120590

=

1

2

(119881119898 minus 119865)119879(119881119898 minus 119865) +

1

2

120582(119877120590)119879(119877120590)2

(18)



Δ120590 =

1199041015840

119903

11990410158401015840119903

=

(1198651015840)

119879

(119881119898 minus 119865) minus 120582(119877)119879(119877120590)

(1198651015840)119879(1198651015840) minus (119865

10158401015840)119879(119881119898 minus 119865) + 120582119877

119879119877

(19)


Δ120590 =

119869119879(119881119898 minus 119891) minus 120582119868120590

119869119879119869 + 120582119868

(20)



[120590119896+1] = [120590119896] + [[119869119879119869 + 120582119868]

minus1

[119869119879[119881119898 minus 119891] minus 120582119868120590]]

119896 (21)











9 Conclusions




References

















































































































































































































































































RoboticsJournal of





Journal of



RotatingMachinery





VLSI Design



Shock and Vibration







Journal of



Volume 2014


SensorsJournal of





Propagation










120597Ω

Ω

VV

V

VV

I













I

V

V

V

V

V

V

V

V

V

VVV

V

1

2

3456

7

8

9

10

1112 13 14

15

16

(a)

I

V

V

V

V

V

V

V

V

V

VVV

V

1

2

3456

7

8

9

10

1112 13 14

15

16

(b)



























Ω






Mul

tiple

xer m

odul

e

CPU

Phantom withsensors

1

23

456

7

8

9

1011

12 1314

15

16

Y

X

minus025

0

025

minus005 0 005



algorithm

Start

Guess

Stop

CPU

No

Yes


12059001205900

120590kJ(120590k) [Vc]k = Γ(120590k)


Measureddata [Vm]k


ΔV lt 120576

Y

X

minus025

0

025

minus005 0 005








nabla sdot 120590nabla120601 = 0 (14)



Φ = Φ119894 (15)



int

120597Ω

120590

120597Φ

120597119899

=



0 otherwise(16)


[Φ] = [119870 (120590119890)]minus1[119862] (17)




119904119903 =

1

2

1003817100381710038171003817119881119898 minus 119865

1003817100381710038171003817

2+

1

2

120582 119877120590

=

1

2

(119881119898 minus 119865)119879(119881119898 minus 119865) +

1

2

120582(119877120590)119879(119877120590)2

(18)



Δ120590 =

1199041015840

119903

11990410158401015840119903

=

(1198651015840)

119879

(119881119898 minus 119865) minus 120582(119877)119879(119877120590)

(1198651015840)119879(1198651015840) minus (119865

10158401015840)119879(119881119898 minus 119865) + 120582119877

119879119877

(19)


Δ120590 =

119869119879(119881119898 minus 119891) minus 120582119868120590

119869119879119869 + 120582119868

(20)



[120590119896+1] = [120590119896] + [[119869119879119869 + 120582119868]

minus1

[119869119879[119881119898 minus 119891] minus 120582119868120590]]

119896 (21)











9 Conclusions




References

















































































































































































































































































RoboticsJournal of





Journal of



RotatingMachinery





VLSI Design



Shock and Vibration







Journal of



Volume 2014


SensorsJournal of





Propagation










I

V

V

V

V

V

V

V

V

V

VVV

V

1

2

3456

7

8

9

10

1112 13 14

15

16

(a)

I

V

V

V

V

V

V

V

V

V

VVV

V

1

2

3456

7

8

9

10

1112 13 14

15

16

(b)



























Ω






Mul

tiple

xer m

odul

e

CPU

Phantom withsensors

1

23

456

7

8

9

1011

12 1314

15

16

Y

X

minus025

0

025

minus005 0 005



algorithm

Start

Guess

Stop

CPU

No

Yes


12059001205900

120590kJ(120590k) [Vc]k = Γ(120590k)


Measureddata [Vm]k


ΔV lt 120576

Y

X

minus025

0

025

minus005 0 005








nabla sdot 120590nabla120601 = 0 (14)



Φ = Φ119894 (15)



int

120597Ω

120590

120597Φ

120597119899

=



0 otherwise(16)


[Φ] = [119870 (120590119890)]minus1[119862] (17)




119904119903 =

1

2

1003817100381710038171003817119881119898 minus 119865

1003817100381710038171003817

2+

1

2

120582 119877120590

=

1

2

(119881119898 minus 119865)119879(119881119898 minus 119865) +

1

2

120582(119877120590)119879(119877120590)2

(18)



Δ120590 =

1199041015840

119903

11990410158401015840119903

=

(1198651015840)

119879

(119881119898 minus 119865) minus 120582(119877)119879(119877120590)

(1198651015840)119879(1198651015840) minus (119865

10158401015840)119879(119881119898 minus 119865) + 120582119877

119879119877

(19)


Δ120590 =

119869119879(119881119898 minus 119891) minus 120582119868120590

119869119879119869 + 120582119868

(20)



[120590119896+1] = [120590119896] + [[119869119879119869 + 120582119868]

minus1

[119869119879[119881119898 minus 119891] minus 120582119868120590]]

119896 (21)











9 Conclusions




References

















































































































































































































































































RoboticsJournal of





Journal of



RotatingMachinery





VLSI Design



Shock and Vibration







Journal of



Volume 2014


SensorsJournal of





Propagation




























Ω






Mul

tiple

xer m

odul

e

CPU

Phantom withsensors

1

23

456

7

8

9

1011

12 1314

15

16

Y

X

minus025

0

025

minus005 0 005



algorithm

Start

Guess

Stop

CPU

No

Yes


12059001205900

120590kJ(120590k) [Vc]k = Γ(120590k)


Measureddata [Vm]k


ΔV lt 120576

Y

X

minus025

0

025

minus005 0 005








nabla sdot 120590nabla120601 = 0 (14)



Φ = Φ119894 (15)



int

120597Ω

120590

120597Φ

120597119899

=



0 otherwise(16)


[Φ] = [119870 (120590119890)]minus1[119862] (17)




119904119903 =

1

2

1003817100381710038171003817119881119898 minus 119865

1003817100381710038171003817

2+

1

2

120582 119877120590

=

1

2

(119881119898 minus 119865)119879(119881119898 minus 119865) +

1

2

120582(119877120590)119879(119877120590)2

(18)



Δ120590 =

1199041015840

119903

11990410158401015840119903

=

(1198651015840)

119879

(119881119898 minus 119865) minus 120582(119877)119879(119877120590)

(1198651015840)119879(1198651015840) minus (119865

10158401015840)119879(119881119898 minus 119865) + 120582119877

119879119877

(19)


Δ120590 =

119869119879(119881119898 minus 119891) minus 120582119868120590

119869119879119869 + 120582119868

(20)



[120590119896+1] = [120590119896] + [[119869119879119869 + 120582119868]

minus1

[119869119879[119881119898 minus 119891] minus 120582119868120590]]

119896 (21)











9 Conclusions




References

















































































































































































































































































RoboticsJournal of





Journal of



RotatingMachinery





VLSI Design



Shock and Vibration







Journal of



Volume 2014


SensorsJournal of





Propagation










Ω






Mul

tiple

xer m

odul

e

CPU

Phantom withsensors

1

23

456

7

8

9

1011

12 1314

15

16

Y

X

minus025

0

025

minus005 0 005



algorithm

Start

Guess

Stop

CPU

No

Yes


12059001205900

120590kJ(120590k) [Vc]k = Γ(120590k)


Measureddata [Vm]k


ΔV lt 120576

Y

X

minus025

0

025

minus005 0 005








nabla sdot 120590nabla120601 = 0 (14)



Φ = Φ119894 (15)



int

120597Ω

120590

120597Φ

120597119899

=



0 otherwise(16)


[Φ] = [119870 (120590119890)]minus1[119862] (17)




119904119903 =

1

2

1003817100381710038171003817119881119898 minus 119865

1003817100381710038171003817

2+

1

2

120582 119877120590

=

1

2

(119881119898 minus 119865)119879(119881119898 minus 119865) +

1

2

120582(119877120590)119879(119877120590)2

(18)



Δ120590 =

1199041015840

119903

11990410158401015840119903

=

(1198651015840)

119879

(119881119898 minus 119865) minus 120582(119877)119879(119877120590)

(1198651015840)119879(1198651015840) minus (119865

10158401015840)119879(119881119898 minus 119865) + 120582119877

119879119877

(19)


Δ120590 =

119869119879(119881119898 minus 119891) minus 120582119868120590

119869119879119869 + 120582119868

(20)



[120590119896+1] = [120590119896] + [[119869119879119869 + 120582119868]

minus1

[119869119879[119881119898 minus 119891] minus 120582119868120590]]

119896 (21)











9 Conclusions




References

















































































































































































































































































RoboticsJournal of





Journal of



RotatingMachinery





VLSI Design



Shock and Vibration







Journal of



Volume 2014


SensorsJournal of





Propagation











nabla sdot 120590nabla120601 = 0 (14)



Φ = Φ119894 (15)



int

120597Ω

120590

120597Φ

120597119899

=



0 otherwise(16)


[Φ] = [119870 (120590119890)]minus1[119862] (17)




119904119903 =

1

2

1003817100381710038171003817119881119898 minus 119865

1003817100381710038171003817

2+

1

2

120582 119877120590

=

1

2

(119881119898 minus 119865)119879(119881119898 minus 119865) +

1

2

120582(119877120590)119879(119877120590)2

(18)



Δ120590 =

1199041015840

119903

11990410158401015840119903

=

(1198651015840)

119879

(119881119898 minus 119865) minus 120582(119877)119879(119877120590)

(1198651015840)119879(1198651015840) minus (119865

10158401015840)119879(119881119898 minus 119865) + 120582119877

119879119877

(19)


Δ120590 =

119869119879(119881119898 minus 119891) minus 120582119868120590

119869119879119869 + 120582119868

(20)



[120590119896+1] = [120590119896] + [[119869119879119869 + 120582119868]

minus1

[119869119879[119881119898 minus 119891] minus 120582119868120590]]

119896 (21)











9 Conclusions




References

















































































































































































































































































RoboticsJournal of





Journal of



RotatingMachinery





VLSI Design



Shock and Vibration







Journal of



Volume 2014


SensorsJournal of





Propagation


















9 Conclusions




References

















































































































































































































































































RoboticsJournal of





Journal of



RotatingMachinery





VLSI Design



Shock and Vibration







Journal of



Volume 2014


SensorsJournal of





Propagation









Documents

Review Article Bioelectrical Impedance Methods for Noninvasive Health Monitoring: A Review · 2019. 7. 31. · Review Article Bioelectrical Impedance Methods for Noninvasive Health