Circuit and Signal Processing for Capacitance Measurement of Breast Tissue

Abstract

In previous studies, electrical capacitance tomography has been developed and applied for imaging breast tissue to detect abnormalities within the breast caused by simple cysts, benign tumors and malignant cancers. For accurate screening and early detection of such abnormalities, high sensitive capacitance sensor is necessary. This study proposes a novel capacitance sensor circuit and signal conditioning and processing based on capacitance–voltage circuit for breast tissue measurement. The new sensor circuit design used a biopotential capacitance electrode, capable of measuring capacitance value with a resolution as low as 0.1 fF, a sensitivity of 1.6 V/pF, and linearity of 0.98. The experiment was conducted with a hemisphere 3D sensor 24 electrodes. The experiment strategy is as follow, first the system will be calibrated using network analyzer, secondly experiment using phantom 1, and the third experiment using phantom 2. In the design, we used a reference electrode made from fixed plate to measure capacitance inside the system, which will be used further for compensation against signal fluctuation caused by environmental condition such as humidity, temperature, pressure, etc. As a result, more stable system is achieved. Based on the experiment, the system can detect abnormalities of the human breast which are represented by two phantoms with different condition.

Full text

Copyright © 2015 American Scientific Publishers
All rights reserved
Printed in the United States of America
Article
Advanced Science,
Engineering and Medicine
Vol. 7, 1–6, 2015
www.aspbs.com/asem
Circuit and Signal Processing for Capacitance
Measurement of Breast Tissue
Arba’i Yusuf12∗, S. Harry Sudibyo1, Dodi Sudiana1, Imamul Muttakin2, Marlin R. Baidillah2,
Dilla Nelvo Dasril2, Wahyu Widada2, and Warsito P. Taruno2 3
1Department of Electrical Engineering, University of Indonesia, Depok, Indonesia
2CTECH Labs Edwar Technology, Tangerang, Indonesia
3Department of Physics, University of Indonesia, Depok, Indonesia
In previous studies, electrical capacitance tomography has been developed and applied for imag-
ing breast tissue to detect abnormalities within the breast caused by simple cysts, benign tumors
and malignant cancers. For accurate screening and early detection of such abnormalities, high
sensitive capacitance sensor is necessary. This study proposes a novel capacitance sensor cir-
cuit and signal conditioning and processing based on capacitance–voltage circuit for breast tissue
measurement. The new sensor circuit design used a biopotential capacitance electrode, capable
of measuring capacitance value with a resolution as low as 0.1 fF, a sensitivity of 1.6 V/pF, and
linearity of 0.98. The experiment was conducted with a hemisphere 3D sensor 24 electrodes. The
experiment strategy is as follow, first the system will be calibrated using network analyzer, secondly
experiment using phantom 1, and the third experiment using phantom 2. In the design, we used a
reference electrode made from fixed plate to measure capacitance inside the system, which will be
used further for compensation against signal fluctuation caused by environmental condition such
as humidity, temperature, pressure, etc. As a result, more stable system is achieved. Based on the
experiment, the system can detect abnormalities of the human breast which are represented by
two phantoms with different condition.
Keywords: Capacitance Measurement, Signal Processing, Sensitivity, Breast Tissue, Cancer
Detection, Biopotential Electrodes.
1. INTRODUCTION
Early detection of breast cancers and other abnormalities
within the breast is extremely of importance to get effec-
tive therapy and avoid fatalities. A number of techniques
used widely to detect lump within the breast in medi-
cal practices are Ultrasonography (USG) and Mammog-
raphy as these methods relatively cheap and easy to use
as compared to Magnetic Resonance Imaging (MRI) and
Computed Tomography (CT) scan. The limitation of these
conventional methods for early detection of abnormalities
within the breast is, that in many cases, a tingling sensation
or a present of lumps felt by a patient is not always con-
firmed by USG or Mammography until the lump became
big enough, which is often already too late. Lumps with
sizes under 1 cm in diameter within the breast usually
∗Author to whom correspondence should be addressed.
are difficult to detect. Alternative methods which are fast,
simple to use, high safety, non-radiation and low cost are
still in needs to develop.
Microwave tomographic imaging introduces significant
capability as a new method for breast cancer early detec-
tion based on contrasts between dielectric properties of
healthy versus malignant tissues.1However, its broadband
natures make the implementation expensive and complex
in hardware levels. Recently, electrical impedance spec-
troscopy as a mean of non-invasive medical technique
that performs surface electrical measurement was inves-
tigated for breast imaging. The system utilizes simple
microcontroller-based circuitry to sense the electrical prop-
erty changes in the body to alternating current. The tech-
nique was claimed as a complementary to mammography
and magnetic resonance imaging (MRI) for breast cancer
detection.2Electrical tomography is considered as having
Adv. Sci. Eng. Med. 2015, Vol. 7, No. xx 2164-6627/2015/7/001/006 doi:10.1166/asem.2015.1779 1

Circuit and Signal Processing for Capacitance Measurement of Breast Tissue Yusufetal.
high sensitivity of electrical properties measurement of
breast tissue so that design methods of circuit and system
have been sought continuously.3
In our previous published studies, we have developed
the first usage of electrical capacitance volume tomog-
raphy (ECVT) for breast cancer imaging called ECVT
Breast Imaging.4The technique is based on capacitance
measurement of the breast tissue using a number of capac-
itive sensors arranged surrounding outside of the breast
and image reconstruction using tomography algorithm of
the breast tissue from the measured capacitance data. The
technique is very fast, i.e., less than one second to get
the whole volumetric image of the breast with abnor-
malities inside, non-radiation and low-cost. However, the
data acquisition of the capacitance measurement used in
the imaging system is sometimes prone to stray capaci-
tance that often caused artifacts in the reconstructed image,
and hence false diagnosis. To perform very early and
highly accurate detection of abnormalities, a stray immune
and high sensitive capacitance sensing is needed to be
developed.
Our previous data acquisition system used for breast
imaging has major constrains in the sensitivity and reso-
lution which are only 0.56 V/pF and 0.42 fF, respectively.
Thus, it is still on demand to improve. In this study, we
develop electronic circuit and signal processing for more
accurate and higher sensitive capacitance measurement of
breast tissues. To improve the sensitivity and stability, a
simultaneous ADC and reference electrode are used.
2. SENSING AND SIGNAL
CONDITIONING CIRCUIT
Sensing and signal conditioning circuits are used to con-
vert capacitance signal from electrode to DC signal for fur-
ther process. The circuit is consisting of signal generator,
capacitance-to-voltage (C–V) circuit, and DC restoration
circuit.610
2.1. Biopotential Electrodes
The most important consideration to be made for the sens-
ing of electric signal from human body is biopotential elec-
trode with some requirements such as safety from electric
discharge and high input impedance.78Figure 1(a) repre-
sents a biopotential electrode which consists of electrode
and dielectric. The electrical characteristics of biopoten-
tial electrode generally is nonlinear, therefore performed a
linear approach in its application. In ideal condition, elec-
trodes can be represented by an equivalent circuit as shown
in Figure 1(b) where, Ehc is half-cell potential, Rd and
Cd are components that represent impedances associated
with electrode-dielectric interface and polarization at the
interface. Rsis series resistance associated with resistance
material of the electrode.
The electrode design in this research is fashioned to
measure capacitance inside the breast tissue as a high
Figure 1. (a) Biopotential electrode; (b) equivalent circuit for biopoten-
tial electrode.
dielectric material used in the capacitive bioelectrode to
form a capacitor between the skin and the electrode.
Figure 2 shows the block diagram to sense capacitance of
breast tissue. Breast is surrounded by several bioelectrodes
and connected to electronic circuit and personal computer
(PC). Electronic circuit contains signal generator and sig-
nal processing unit. Signal generator produces sine-wave
signal injected into the bioelectrode as excitation electrode,
while signal processing is converts the capacitance signal
from detection electrode into voltage for further process.
The electric field will be generated between electrodes pair
inside breast, thus the unknown capacitance (Cxinside
breast can be measured. Normal and abnormality of the
breast will produce different electric field pattern and so
does unknown capacitance (Cx. Personal computer func-
tioned to reconstruct all of capacitance data to form an
image. The required 3D ECVT algorithm was explained
in other paper.9
2.2. Capacitance to Voltage Circuit
The capacitance signals from electrodes pairs need to
be converted into voltage for further process using an
electronic circuitry namely capacitance-to-voltage (C–V)
circuit. The C–Vcircuit is built by an operational ampli-
fier, feedback resistor, and feedback capacitor which would
convert the current into ac voltage Vo(Fig. 3). Cqis stray
capacitance imposed by screen sensor, cable, and elec-
tronic switches.10 Kirchhoff law states that current entering
node and leaving node is equal (i1=i2, thus, the voltage
Figure 2. Block diagram of breast tissue measurement.
2Adv. Sci. Eng. Med. 7, 1–6,2015

Yus u f e t a l . Circuit and Signal Processing for Capacitance Measurement of Breast Tissue
Figure 3. Basic of C–Vcircuit.
representing capacitance measurement can be retrieved
from:
i1=i2(1)
Vi
Zx
=−Vo
Zf
(2)
Vi
1/jCx=− Vo
Rf1/jCf/Rf+1/jCf (3)
Vo=− RfViCx/Cf
Rf+1/jCf(4)
k=Rf
Rf+1/jCf(5)
where Vois voltage output that represents capacitance mea-
surement, Viis sinusoidal voltage input injected to the
electrode, Rfis feedback resistor, Cfis feedback capacitor,
is angular frequency, Cxis object capacitance to be mea-
sured. This equation contains two elements i.e., capacitive
and resistive.
The smaller resistance value causes greater conductivity
and vice versa. In the capacitance measurement method,
conductivity will be damped by setting k≈1, which can be
made by selecting suitable value of Rfand Cf. Therefore
the electronic circuit only measures capacitance value, and
Eq. (4) can be simplified to:
Vo Cx
Cf
Vi(6)
DDS (direct digital synthesizer) is used in this system
as excitation source that produces continuous sine wave
with its frequency which can be set from 0 Hz to 50 MHz.
In addition, on the output of C–Vcircuit there will be
bipolar wave output voltage with positive–negative value,
which its peaks of signal representing capacitance mea-
surement. The signal must be converted into DC (direct
current) signal using high-speed peak detector sample-
and-hold for further processing and reading by analog-to-
digital conversion (ADC).11
3. SIGNAL PROCESSING
Signal processing is used to analyze and convert certain
measured signal into another information. In this paper,
signal must be converted into capacitance that represents
capacitance of breast tissue. Figure 4 shows block diagram
of signal processing, which consists of preconditioning
(C–Vcircuit, initial gain, peak detector, and low pass fil-
ter) as mentioned above to convert the capacitance signals
from electrodes pairs into voltages, simultaneous ADC,
microcontroller, reference electrode, and serial USB. The
reference electrode is a fixed plate used to measure capac-
itance inside the system, which will be used further for
compensation against signal fluctuation caused by envi-
ronmental condition such as humidity, temperature, pres-
sure, etc.
3.1. Analog-to-Digital Conversion and Microcontroller
Analog signal from DC-restoration circuit needs to be con-
verted into digital value using simultaneous ADC AD7606
from analog device. This chip has features such as 16-bit
resolution, 200 KHz sampling rate for all channel, bipolar
analog input range, voltage reference internal, and paral-
lel output. The capacitance from all electrodes pair and be
read simultaneously, thus no delay between channels.
The microcontroller used for acquiring data from simul-
taneous ADC, calculation into capacitance value, and send
it to personal computer.
3.2. Capacitance Measurement
Capacitance to be measured between pair electrodes are
not exactly Cx, but there is another parasitic capacitance
or stray capacitance measured in parallel with Cx.Total
capacitance Cx+Cqis named standing capacitance Cs.
Stray capacitance Cqcan be measured when no object
between pair electrode, hence the system only measure
screen guard, cable, and electronic switches and then
save it to register memory inside microcontroller. Capac-
itance Cxis obtained by subtracting standing capaci-
tance and stray capacitance. The calculation procedures are
described as:
CsC
q=VAD
G1V1
Cf(7)
Cx=Cs−Cq(8)
where Csis standing capacitance (Cxinfluenced by stray
capacitance Cq), VAD is voltage by ADC reading, Cfis
capacitance feedback, Gis initial gain inside C–Vcircuit,
and Cxis measured capacitance.
4. EXPERIMENTAL RESULT
The experiment was conducted with a hemisphere 3D sen-
sor 24 electrodes as depicted in Figure 5 that divided
into three levels, each level contains eight electrode.5
The experiment strategy is as follow: first the system
Adv. Sci. Eng. Med. 7, 1–6, 2015 3

Circuit and Signal Processing for Capacitance Measurement of Breast Tissue Yusufetal.
Figure 4. Block diagram of signal processing.
will calibrated using network analyser, secondly experi-
ment using phantom 1, and the third experiment using
phantom 2.
4.1. System Calibration
As initial test, two-channel capacitance measurement took
place. Ceramic capacitor from 1 pF until 100 pF were
used to observe resolution, sensitivity and linearity of the
measurement module.
Having examined the noise level, resolution was cal-
culated 0.1 fF with sensitivity of 1.6 V/pF. Compared
with Precision Network Analyzer (PNA Agilent N5221A)
Figure 5. A hemisphere 3D sensor 24 channel: Design with unit dimensions are in meter (left), physical construction with shielding case and ports
(right).
measurement, linearity of the module was plotted in
Figure 6 so as to introduce 2.97% error.
4.2. Experiment Using Phantom 1
Phantom in the experiment was used to describe abnormal-
ities of the human breast. Phantom was made from paraffin
(r=22) which represented fat inside the breast and small
ball made from natrium chlorida (NaCl, r=45) with dif-
ferent composition to model abnormalities of breast tissue
such as cancer or another lump.
Table I shows reconstruction result of phantom based
on ECVT image reconstruction technique described
4Adv. Sci. Eng. Med. 7, 1–6,2015

Yus u f e t a l . Circuit and Signal Processing for Capacitance Measurement of Breast Tissue
0 10 20 30 40 50 60 70 80 90 100
0
10
20
30
40
50
60
70
80
90
100
Capacitance measurement with PNA (pf)
Capacitance measurement module (pf)
Figure 6. Comparison of measurement capacitor between system
and PNA.
elsewhere.459The table shows in axial and lateral images
using phantom with diameter 3.5 cm. For phantom detec-
tion, capacitance measurement is normalized to obtain rel-
ative capacitance using equation:
dn =Cr−Cl
Ch−Cl
(9)
Tab l e I . Detection result using phantom 1.
Phantom Axial images Lateral Images
NaCl 3gr
NaCl 13gr
NaCl 23gr
NaCl 13gr and
23gr
Table II. Value of normalize absolute capacitance at phantom 1.
Sodium Concentrate Detected Detected Normalize
level sodium volume diameter absolute
Phantom (gr) (gr/cm3(cm3(cm) capacitance
130134 14795 6.561 0.0526
213058 23603 7.667 0.0964
3231025 245442 7.767 0.1293
4 13 and 23 1604 193346 7.174 0.1699
where dn is normalization result of absolute capacitance,
Cris measured capacitance, Cland Chare capacitance
at empty calibration (sensor filled with low permittivity
material) and full calibration (sensor filled with high per-
mittivity material) respectively. Table II shows value of
normalized absolute capacitance at phantom.
4.3. Experiment Using Phantom 2
The next experiment is using phantom 2 composed a wet
paper tissue (r=80) which represented skin of the breast
and small ball of NaCl (r=45) to model abnormalities of
breast tissue such as cancer or another lump. The method
is slightly different with phantom 1. When using phan-
tom 1 the image can be shown directly because the small
ball of NaCl is surrounded by paraffin with low permit-
tivity, while using phantom 2 the image can’t be shown
directly because the small ball of NaCl is surrounded by
wet paper tissue with high permittivity. The abnormali-
ties of human breast (Ancp) can be detected by means of
subtraction the image (Tncp) with normal image of breast
which is represented by wet paper tissue (Wncp).
Ancp =Tncp −Wncp (10)
Table III. Detection result using phantom 2.
Ncp Images
(1)Wet Paper
Tissue (Wncp)
(2)Wet Paper
Tissue + NaCl (Tncp)
Result of
abnormalities (Ancp)
Adv. Sci. Eng. Med. 7, 1–6, 2015 5

Circuit and Signal Processing for Capacitance Measurement of Breast Tissue Yusufetal.
5. CONCLUSIONS
Circuit and signal processing for capacitance measurement
of breast tissue has been design and assembled. The sys-
tem uses sine-wave excitation with frequency range can be
adjusted from 0 KHz to 50 MHz and capable of measuring
capacitance change with resolution 0.1 fF and 1.6 V/pF of
sensitivity. Comparison with Precision Network Analyser
(PNA) shows that the system is sufficiently accurate to
measure capacitance with 0.98 of linearity. In the design,
we used a reference electrode made from fixed plate to
measure capacitance inside the system, which will be used
further for compensation against signal fluctuation caused
by environmental condition such as humidity, temperature,
pressure, etc. As a result, more stable system is achieved.
Based on the experiment, the system can detect abnormali-
ties of the human breast represented by two phantoms with
different condition.
References and Notes
1. N. Irishina, M. Moscoso, and O. Dorn, IEEE Transactions on
Biomedical Engineering 56, 1143 (2009).
2. E. Alzurq, A. Almaktari, B. S. Aldin, M. Hamoud, S. Othman, and
S. Nasr, Recent Advances in Environmental Science and Biomedicine
142 (2014).
3. S. Hong, K. Lee, U. Ha, H. Kim, Y. Lee, Y. Kim, and H.-J. Yoo,
IEEE Journal of Solid-State Circuits 50, 245 (2015).
4. W. P. Taruno, M. R. Baidillah, R. I. Sulaiman, I. Widora, A. Yusuf,
W. S. Widada, and M. Aljohani, Adv. Sci. Eng. Med. 6, 1
(2014).
5. W. P. Taruno, M. R. Baidillah, R. I. Sulaiman, A. Yusuf, W. Widada,
H. Alzufri, and M. Aljohani, A novel sensor design for breast
cancer scanner based on electrical capacitance volume tomography
(ECVT), IEEE Sensor 2012 Conference, Taipei, Taiwan, October
(2012), pp. 96–99.
6. A. Yusuf, I. Muttakin, R. A. Rohmadi, W. Widada, and W. P. Taruno,
Single signal conditioning multi electrode for ECVT data acquisition
system, TENCON, October (2014).
7. D. Prutchi and M. Norris, Design and Development of Medical Elec-
tronics Instrumentation, Wiley-Intersicence, A Jhon Wiley & Son
Inc Publication (2004), pp. 1–40.
8. M. R. Neuman, The Biomedical Engineering Handbook: Biopoten-
tial Electrodes, CRC Press LLC (2000),p.48.
9. Warsito, Q. Marashdeh, and L. S. Fan, IEEE Sensor Journal 7, 525
(2007).
10. W. Q. Yang, Meas. Sci. Technol. 7, 225 (1996).
11. A. Yusuf, I. Muttakin, W. Widada, and W. P. Taruno, Analysis of sin-
gle excitation signal for high speed ECVT data acquisition system,
ICITEE, October (2014), pp. 360–365.
Received: 10 October 2014. Accepted: 15 November 2014.
6Adv. Sci. Eng. Med. 7, 1–6,2015