Hindawi

International Journal of Biomedical Imaging

Volume 2018, Article ID 8241438, 12 pages
https://doi.org/10.1155/2018/8241438

Research Article
EBG Based Microstrip Patch Antenna for
Brain Tumor Detection via Scattering Parameters in
Microwave Imaging System

Reefat Inum , Md. Masud Rana, Kamrun Nahar Shushama, and Md. Anwarul Quader
Department of Electrical and Electronic Engineering, Rajshahi University of Engineering and Technology, Rajshahi, Bangladesh

Correspondence should be addressed to Reefat Inum; [email protected]

Received 5 May 2017; Revised 10 November 2017; Accepted 25 December 2017; Published 12 February 2018

Academic Editor: Tiange Zhuang

Copyright © 2018 Reefat Inum et al. This is an open access article distributed under the Creative Commons Attribution License,
which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

A microwave brain imaging system model is envisaged to detect and visualize tumor inside the human brain. A compact and efficient
microstrip patch antenna is used in the imaging technique to transmit equivalent signal and receive backscattering signal from the
stratified human head model. Electromagnetic band gap (EBG) structure is incorporated on the antenna ground plane to enhance
the performance. Rectangular and circular EBG structures are proposed to investigate the antenna performance. Incorporation of
circular EBG on the antenna ground plane provides an improvement of 22.77% in return loss, 5.84% in impedance bandwidth,
and 16.53% in antenna gain with respect to the patch antenna with rectangular EBG. The simulation results obtained from CST are
compared to those obtained from HFSS to validate the design. Specific absorption rate (SAR) of the modeled head tissue for the
proposed antenna is determined. Different SAR values are compared with the established standard SAR limit to provide a safety
regulation of the imaging system. A monostatic radar-based confocal microwave imaging algorithm is applied to generate the image
of tumor inside a six-layer human head phantom model. 𝑆-parameter signals obtained from circular EBG loaded patch antenna
in different scanning modes are utilized in the imaging algorithm to effectively produce a high-resolution image which reliably
indicates the presence of tumor inside human brain.

1. Introduction recently as it offers a safe, rapid, low-cost, noninvasive, and

highly accurate system solution which involves nonionizing
Brain cancer is one of the serious public health problems radiation [3].
worldwide because it affects the most vital organ of human Microwave imaging is an active wave-based noninvasive
body. For example, in the USA, 23,800 patients and 16,700 imaging method. Nonionizing electromagnetic waves from
deaths are estimated due to brain cancer in 2017 [1]. Such microwave signals are able to penetrate human tissues with-
a high death rate is caused by the invasive properties of out creating health hazards [4]. The contrast in the electrical
tumors which turn brain cancer into a serious disease. But it properties between healthy and malignant tissues is the
is encouraging that the cure rate can be increased by reliably principal of operation of microwave imaging systems. Four
diagnosing it in the early stages because treatment at early large groups, namely, optimization-based microwave imag-
stage is more efficient and effective compared with treatment ing, microwave tomography, confocal radar-based imaging,
done at the late stage of cancer. The common imaging and microwave holography can be considered as the branches
modalities utilized to detect cancer are magnetic resonance of currently active microwave systems for tissue imaging
imaging (MRI) scanning, X-ray screening, computed tomog- [5]. Radar-based techniques are preferable since they only
raphy (CT) scans, positron emission tomography (PET), focus on detecting the tumor rather than the entire range
and ultrasound imaging [2]. The possibility of microwave of electrical properties [6]. Therefore, much easier signal
imaging technology for brain cancer detection is increased processing such as less sophisticated delay and sum confocal
2 International Journal of Biomedical Imaging

Human head phantom model

Antenna

Microwave Signal storing and

transceiver processing unit

Figure 1: Schematic diagram of monostatic radar-based microwave brain imaging system [9].

microwave imaging algorithm is involved in radar-based The goal of this study is twofold: (1) design verification
microwave imaging. of an EBG based microstrip patch antenna by comparing the
In any brain imaging system, the antennas form an results obtained from CST and HFSS software and (2) detec-
important element for the quality of the final image and tion of tumor inside human brain by radar-based microwave
the performance of imaging. The appropriate antenna for brain imaging system. In [9], the simulation results obtained
one such system should cover a number of requirements from CST showed that the patch antenna with circular EBG
including ease of integration, compactness, simple geometric structure provides best performance and consequently it is
structure, small dimensions, enhanced bandwidth, gain, and applied in the brain tumor detection application. The normal
directionality [7–11]. These requirements can be fulfilled by six-layered spherical human head phantom modeled in [9]
a microstrip patch antenna because it possesses a number of is made, a cancerous phantom by inserting a tumor model
advantages like low profile, light weight, low volume, planar inside head. The dimensions and electrical properties of
configuration, low fabrication cost, and so forth [12]. How- head and tumor models are chosen feasibly that reflect the
ever, conventional patch antennas need some modifications
actual ones. To get the imaging results, proposed antenna is
for reliable pulse transmission and collection of backscattered
simulated with head phantom model by CST MWS software.
signals in microwave imaging systems. Amalgamation of
The 𝑆-parameter results obtained from different scanning
electromagnetic band gap (EBG) structure on the antenna
ground plane is found in the literature to be an effective way modes involving head phantom with and without tumor
to enhance the performance of conventional microstrip patch model are applied in the microwave imaging algorithm to
antenna [13]. EBG structures are any artificial periodic objects reconstruct tumor image. The SAR values for three different
designed to prevent/assist electromagnetic wave propagation measuring techniques are compared with each other and with
in a given band of frequency. Due to their unique band the established standard SAR limit so that safety of microwave
gap features, EBG structures can be categorized as a special brain tumor detection system is ensured. Thus, EBG based
type of metamaterial. This unique property has been applied microstrip patch antenna is applied in brain tumor detection
to design antenna systems with better gain and efficiency, for the first time with all the necessary aspects of microwave
reduced mutual coupling, and lower side-lobes and back-lobe brain imaging system being clearly analyzed.
levels by suppressing surface wave modes [14, 15].
Assessment of specific absorption rate (SAR) induced
inside human head is another important feature of any
2. Radar-Based Microwave Brain Imaging
microwave brain imaging system. EM waves radiated from System Model
the transmitting antenna directly travel through the patients
The aim of radar-based imaging method is to use backscat-
head and significant portion of the radiated power carrying
tering signals for the creation of images which can detect
by EM waves is absorbed by head tissues. These energy
the presence of an object embedded in a dielectric medium.
absorptions are not distributive in nature and may cause
localized RF energy deposition in the form of nodes and The received signal is stored and a suitable signal processing
damage DNA of living tissues which may trigger cell suicide algorithm makes the image possible. Figure 1 depicts the
or unregulated cell division resulting in the formation of a concept of our considered brain imaging system [9]. In this
cancerous tumor [16]. Hence accurate SAR analysis which system model, we specifically focus on the effective antenna
greatly depends on the exact head phantom model is a modeling, human head phantom modeling, detection of
must one in imaging system to ensure the safety of patient tumor inside human brain by radar-based microwave image
under test. As the precise modeling of multilayered human reconstruction algorithm, and determination of SAR values
head consisting of different complex tissues is cumbersome, inside six-layer human head model.
various head phantom models have been investigated to The head-imaging architecture shown in Figure 1 primar-
mimic the practical scenario [17–19]. ily consists of a compact antenna to transmit and receive
International Journal of Biomedical Imaging 3

wideband signals, a microwave transceiver for signal genera- Table 1: Rectangular microstrip patch antenna specifications.
tion, and data acquisition and a personal computer for signal
Optimized
processing and image formation. A head-imaging platform Dimensions
Design parameters dimensions with
is designed virtually in CST MWS software to evaluate the without EBG (mm)
circular EBG (mm)
system’s performance on detecting tumor using a realistic
Patch width, 𝑊 14.53 14.53
head phantom. The head phantom is scanned by changing
the antenna position in different manner and the reflected Patch length, 𝐿 11.53 8.95
signals are collected and converted from frequency domain Inset feed, 𝑦0 3.125 3.41
to time domain. The resulting signals are then processed and Inset gap, 𝑔 1.901 0.38
used in confocal microwave image reconstruction algorithm Feed length, 𝐿 𝑓 13.202 11.895
to visualize the presence of tumor inside human head. Feed width, 𝑊𝑓 1.901 1.9010
Substrate width, 𝑊𝑆 31.02 31.0170
2.1. Antenna and EBG Structure Design. The development of Substrate length, 𝐿 𝑆 31.68 25.92
microstrip patch antenna with EBG structure and relevant
Substrate thickness, ℎ 0.75 1.03
theory is discussed in this section. Initially, a rectangular
microstrip patch antenna is designed without any EBG
structure and taking 7.3 GHz as the resonant frequency. obtain both the permeability and permittivity from the 𝑆-
The reason behind the choice of such working frequency parameter. The relative material parameters can be calculated
is twofold. Firstly, higher frequency helps to improve the as follows:
spatial resolution of the reconstructed tumor image and,
𝑐 2 𝜔 2
secondly, relatively larger resonant frequency can ensure a 𝑛2 = 𝜀𝑟 𝜇𝑟 = − [ ln (𝑒𝛾𝑑 )] + ( 𝑐 ) ,
compact antenna with small dimension, which is important 𝜔𝑑 𝜔
for any brain imaging system. The antenna is designed (1)
2 2
on Rogers R03003 substrate with relative permittivity of 1 + Γ 𝑛 − (𝜔𝑐 /𝜔)
𝜇𝑟 = √ ,
3 and with the thickness and dimension of 0.75 mm and 1 − Γ 1 − (𝜔𝑐 /𝜔)2
31.68 mm × 31.02 mm, respectively. Due to its flexibility for
high frequency application compared to FR4, Rogers R03003 where 𝑛 is the refractive index, 𝜀𝑟 is the relative permittivity,
is chosen as the substrate material [9]. However, the perfor- 𝜇𝑟 is the relative permeability, 𝑑 is the maximum length
mance of an antenna is greatly affected by substrate thickness of metamaterial unit cell, 𝑐 = 1/√𝜀0 𝜇0 is the speed of
and it is known that a thicker substrate can enhance the light in free space, 𝜔 is the angular frequency, and 𝜔𝑐 is
antenna efficiency and bandwidth but in exchange a surface the angular cutoff frequency of the incident wave on the
wave is initiated which reduces the amount of radiated power metamaterial. The inverse to the quantity 𝑒𝛾𝑑 is known as
[20]. the phase factor for electromagnetic (EM) wave propagating
Figure 3 shows the geometry of the patch antenna without through the metamaterial and is given by
EBG structure, where top view of the antenna is illustrated
with all necessary dimension parameters. The detail design 𝑆11 + 𝑆21 − Γ
𝑒−𝛾𝑑 = , (2)
procedure with all the essential equations is available in 1 − (𝑆11 + 𝑆21 ) Γ
[20]. Microstrip line feeding technique is used to feed the
antenna and 50 Ω is taken as the characteristic impedance where 𝛾 = 𝑗𝜔√𝜀0 𝜇0 √𝑛2 − (𝜔𝑐 /𝜔)2 is the propagation constant
of the transmission line. Next, two different types of EBG and Γ is given by
structures are proposed to improve the antenna performance.
Incorporation of EBG structures on the antenna ground
Γ = 𝑋 ± √𝑋2 − 1, (3)
plane results in a reduction of patch length, substrate length,
inset gap, and an increase in the substrate height. Table 1 2 2
𝑆11 − 𝑆21 +1
shows all the dimension parameters to design the patch 𝑋= , (4)
2𝑆11
antenna without EBG and optimized antenna with circular
EBG structure. The geometry of the patch antenna with where 𝑆11 and 𝑆21 are the reflection and transmission coef-
EBG structure is shown in Figure 4(a). The dimensions of ficients which are collected by simulating the metamaterial
EBG structures are determined from the wavelength of the unit cell. These reflection and transmission coefficients are
microwave signal radiated by the proposed antenna. Each utilized in the above equations to obtain the permittivity
side of the rectangular EBG and diameter of the circular EBG and permeability curves as shown in Figures 5(a) and 5(b).
are calculated as 0.1𝜆, as can be seen in Figures 4(b) and These figures show that the designed EBG structures work
4(c), respectively. The distance between two adjacent EBGs as metamaterial as they provide negative permittivity and
in either case is determined as 0.02𝜆, which is evident from permeability in the desired frequency of 7.3 GHz. Figure 5(a)
Figures 4(b) and 4(c). shows the extracted material parameters (𝜀 and 𝜇) for the cir-
In designing the proposed EBGs, Nicholson-Ross-Weir cular EBG structure, which indicate that both the permittivity
(NRW) technique [21] is used as a conversion approach to and permeability are negative in the frequency range of 3 GHz
4 International Journal of Biomedical Imaging

Table 2: Dielectric properties of the spherical cancerous head proposed microstrip patch antenna is located at 10 mm
phantom with approximate dimension. distance from the phantom surface for suitable penetration
of microwave signal inside the tissue layers.
Radius Permittivity Conductivity, 𝜎
Tissue
(mm) (𝜖𝑟 ) (S/m)
Brain 81 43.22 1.29 3. Results and Discussions
CSF 83 70.1 2.3 The performance of the designed antennas is initially eval-
Dura 83.5 46 0.9 uated using Finite Integration Technique (FIT) based CST
Bone 87.6 5.6 0.03 Microwave Studio (MWS) software and later verified by
Fat 89 5.54 0.04 Finite Element Method (FEM) based HFSS software. Reflec-
Skin 90 45 0.73 tion coefficient, gain, and bandwidth are considered as
the antenna performance parameters. After analyzing the
Tumor 5 55 7
antenna performance from microwave imaging perspectives,
backscattering signals are collected and processed by confocal
microwave imaging algorithm to reconstruct tumor image.
to 8 GHz. For the rectangular EBG structure, the extracted Finally biocompatibility analysis is conducted by assessment
parameters (𝜀 and 𝜇) are negative in the frequency range of of SAR induced inside the cancerous head to determine
6.6 GHz to 7.8 GHz, as evident from Figure 5(b). In both the whether the considered imaging system is prone to health
cases, the antenna resonance frequency (7.3 GHz) falls in the hazard or not.
range of frequency for which the extracted permittivity and
permeability are negative. Therefore, the proposed rectangu-
lar and circular EBG structures are suitable for incorporation 3.1. Performance Analysis of the Designed Antennas. For
on the ground plane of the designed patch antenna and are better impedance matching, the reflection coefficient of an
expected to improve the antenna performance by sufficiently antenna must be at least −10 dB. Specifically, for the antennas
reducing the unwanted surface wave. to be applied in a brain imaging system, the reflection coeffi-
cient should be more negative to ensure minimal reflection of
the input power from antenna to the source. The resolution
2.2. Modeling of Cancerous Human Head Phantom. The of the reconstructed image in microwave imaging depends
frequency dependent dielectric properties (permittivity and on the scattered signal power which is proportional to the
conductivity) of the designed head phantom must imitate the radiated power [9]. The amount of total radiated power from
properties of real human head so that the interaction between the scanning antenna depends on the input power which
EM waves and head tissues can be correctly analyzed. For the will be maximum if the reflection coefficient is enhanced.
radar-based microwave brain imaging, the phantom design Therefore, the quality of the reconstructed image can be
can be simplified in an approximation as a multilayer model. somewhat improved by enhancing the reflection coefficient
In [9], a six-layered spherical head phantom that mimics of the scanning antenna. The performances of the designed
the realistic human head is modeled for the measurement of antennas in terms of reflection coefficients are depicted in
specific absorption rate (SAR). In this paper, the healthy head Figure 8.
phantom in [9] is made cancerous by inserting a tumor inside It is observed from Figures 8(a), 8(b), and 8(c) that the
the brain, as shown by a schematic representation in Figure 6. reflection coefficients obtained from CST closely agree with
A human brain model with tumor inside it, as shown in that obtained from HFSS, for the patch antenna without EBG,
Figure 6, is used in the simulation. The tumor is placed in with rectangular EBG, and with circular EBG, respectively.
between CSF and dura close to the bone. The radiated signals Figure 8(d) compares the reflection coefficients for three
from the antenna start penetrating the outer tissue layer and antennas, which indicate that the normal patch antenna
gradually move toward the inner layer. However, attenuation and patch antenna with rectangular EBG and with circular
of RF energy occurs very quickly due to backscatter from each EBG provide reflection coefficient of −18.40 dB, −40.15 dB,
tissue layer. Therefore, the tumor is placed close to the bone
and −49.29 dB, respectively. Therefore, the incorporation of
so that it can be easily detected from the simulation results. In
circular EBG structure on the antenna ground plane offers the
contrast, detection of heavily buried tumor inside multilayer
best result in terms of reflection coefficient. The information
human brain becomes difficult as the reflected signal from
tumor in this case could be too weak. The dimension and on impedance bandwidth for the three antennas can also be
dielectric properties of all head tissues and tumor model are obtained from Figure 8, which show that the patch antennas
given in Table 2 [17, 18]. without EBG, with rectangular EBG, and with circular EBG
Considering the above realistic dielectric tissue prop- provide impedance bandwidth of 156.20 MHz, 275.50 MHz,
erties in the frequency band of interest, tumor and a six- and 291.60 MHZ, respectively. Hence in this case also,
layered heterogeneous head phantom are modeled to simplify antenna with circular EBG gives the best performance among
the simulation study. In monostatic radar-based microwave other designs.
imaging technique, a single antenna configuration is used to Besides impedance bandwidth, the performance of the
scan the brain phantom with backscattering signal recorded designed antennas is also investigated in terms of gain
at each antenna position. Figure 7 illustrates the simulation bandwidth. Figure 9 shows the gain of the designed antennas
setup for the considered brain imaging method where the as a function of frequency. It is observed from Figure 9 that
International Journal of Biomedical Imaging 5

Table 3: Performance comparison of the proposed antenna with other existing antennas.

Parameter Proposed antenna [10] [11] [14] [22] [23]

Patch dimension (mm2 ) 14.53 × 8.95 50 × 37.5 25 × 25 28 × 28 14 × 14 15.25 × 9.07
Reflection coefficient (dB) −49.289 NR −20 −22 −18 −17.50
Gain (dBi) 6.77 2.6 6.6 2.2 2.5 5.01
Bandwidth (GHz) 0.29 2.16 0.26 0.03 0.19 2.89
NR: not reported.

the patch antenna with circular EBG provides a flat gain of Finally, a comparison between the performance param-
approximately 6.7 dBi from 6 to 8 GHz, while the normal eters of the proposed antenna and similar other existing
patch antennas without EBG and antenna with rectangular antennas is given in Table 3. Here, microstrip patch antenna
EBG provide approximately a flat gain of 6.06 dBi (6.4 to with circular EBG structure is proposed for the considered
7.4 GHz) and 5.89 dBi (6.3 to 7.4 GHz), respectively. It can brain imaging application, due to its improved performances
be noted that the working frequency (7.3 GHz) falls in the as evident from the above discussion. Similar other anten-
range of frequencies over which all the three antennas provide nas mean those designed incorporating any metamaterial
a flat gain bandwidth. Also, the antenna with circular EBG structure [14, 22, 23] or antennas designed for brain imaging
structure offers the largest flat gain bandwidth in comparison application [10, 11]. It is obvious from Table 3 that the pro-
to the other two antennas. Therefore, signal transmission posed microstrip patch antenna with circular EBG structure
and reception with reduced distortion and dispersion can possesses a compact size in comparison to other antennas,
be attained by the circular EBG based antenna which makes which is extremely beneficial for microwave brain imaging
it a potential candidate for the envisaged brain imaging system. It is also evident from the comparison table that the
application. antenna proposed in this paper outperforms similar other
The 𝐸-plane (𝜑 = 90∘ ) radiation patterns of the designed antennas with respect to gain and reflection coefficient. In
antennas in terms of gain are shown in Figure 10(a). It the case of bandwidth, the proposed antenna offers somewhat
is evident from this figure that there is a close agreement little impedance bandwidth of 0.29 GHz, the reason of which
between the 𝐸-plane patterns obtained from CST and HFSS, may be the use of a very thin (35 𝜇m) conductor. However,
which validate the design. Figure 10(a) displays that the the antenna designed in [11] has been utilized in head-
normal patch antenna without EBG and antenna with rectan- imaging application, with a measured impedance bandwidth
gular EBG and with circular EBG provide a gain of 6.17 dBi, of 0.26 GHz. Therefore, circular EBG based patch antenna
5.91 dBi, and 6.77 dBi, respectively, at 7.3 GHz. The main lobe proposed in this paper can be reliably applied in microwave
of the 𝐸-plane pattern in terms of enhanced gain for antenna brain imaging, with an impedance bandwidth of 0.29 GHz.
with circular EBG is directed at 𝜃 = 20∘ . Radiation charac-
teristics like 3 dB beamwidth and side-lobe level for circular 3.2. Brain Tumor Detection. A sequence of task is done as
EBG based patch antenna are 92.6 degree and −8.9 dB. The depicted in the algorithm in Figure 2 to detect the tumor
enhancement in the radiation characteristics for antenna inside head phantom using a monostatic radar-based con-
with circular EBG structure is particularly important from focal microwave image reconstruction system. Monostatic
imaging perspectives, because it ensures reduced distortion radar-based technique adopted in this work is based on the
and dispersion in signal transmission and reception and concept where a single antenna scans the head phantom by
increased resolution capabilities of the scanning antenna.
mechanically rotating around it. However, in CST simulation
Figure 10(b) shows the 𝐻-plane (𝜑 = 0∘ ) radiation
environment, mechanical rotation is accomplished by mov-
patterns in terms of gain at 7.3 GHz for the three antennas.
ing the antenna to all the possible locations around the phan-
It is obvious from this figure that the patterns obtained from
CST and HFSS closely agree with each other for all the three tom and simulating the system for each antenna location.
designs, which further indicate a validation of the design. A Figure 11 illustrates the antenna positions around the head
comparison of the 𝐻-plane patterns between three antennas phantom in different scanning mode. Figures 11(a) and 11(c)
is also provided in Figure 10(b) which reveals that the antenna show the scanning in 𝑥-𝑦 and 𝑥-𝑧 plane. To get a clearer visu-
with circular EBG performs the best (6.36 dBi) in comparison alization of the 3D system in 2D plane, Figures 11(b) and 11(d)
to the antenna without EBG (5.92 dBi) and with rectangular provide scanning performance in reverse 𝑥-𝑦 and reverse
EBG (5.17 dBi). It is also observed from Figure 10(b) that the 𝑥-𝑧 plane. In all these scanning modes, multiple reflections
main lobe direction for all the three antennas is at 𝜃 = 0∘ , from different head tissue layers and tumor response, and
which is parallel to the 𝑧-axis and thereby facilitates posi- noise are contained in the collected scattering signals. As
tioning the antenna in the scanning mechanism. However, the microwave signal reaches tumor after penetrating some
the improved gain offered for the circular EBG based patch complex tissue layers, sufficient signal attenuation occurs
antenna in both the planes is a consequence of the unique which make tumor response quite weak resulting in an easy
structural feature of circular shaped EBG which strongly drowning out in the noise. Therefore, signal processing is a
reduces the surface waves produced on the antenna substrate. must one to remove all the unwanted signals which results
6 International Journal of Biomedical Imaging

Signal transmission from antenna to head phantom and reflection from air-skin interface

S-parameter data acquisition

Frequency to time domain conversion by inverse Fourier transformation

Calibration to get reflection from tumor by subtracting method

Data processing including smoothing filter technique and synthetic focusing estimation

Generation and reconstruction of image

Figure 2: Confocal microwave image reconstruction algorithm for the considered brain imaging system model.

Ws twice the time taken by incident EM wave to pass through

the surface of the object [24]. Figure 12(b) shows that the
Wf energy reflected from skin layer of head phantom with tumor
Lf is greater than that without tumor. The strong reflection from
g the high permittivity tumor tissue is the reason behind this
y0
result. Enlarged portion of tumor response is illustrated in
L Ls Figure 12(c) which shows that reflected transient signal for
head phantom without tumor becomes steady state quicker
W
than that with tumor.
From the above discussion it is clear that the reflected
energy depends on the presence of tumor which in turn
depends on antenna position. This basic principle is used
Figure 3: Schematic diagram of patch antenna without EBG to generate microwave images by moving the antenna to
structure. different locations surrounding the spherical human head
phantom, with scattering signals recorded at each stop
location. The images are created by using the intensities of
in a high-resolution image with low noise level for reliable the scattering signals at each scan position. The resulting
diagnosis. A reference transient scattering signal using a head microwave image of tumor inside multilayer human head
phantom without tumor being presence is needed to remove phantom is shown in Figure 13. Presence of tumor with
all unwanted parts of signal. By subtracting the reference two false points can be observed in Figure 13(a) where
signal from originally received signal, tumor response can scanning is performed in 𝑥-𝑦 plane. Due to reflections from
be extracted. To obtain the reference and original signals, different tissue layers in a cylindrical scanning mode in 𝑥-
proposed antenna is simulated using head phantom with 𝑦 plane, presence of two false points and a nonspherical
and without tumor. Reflection coefficient for original and shaped tumor is evident from Figure 13(a). When the same
reference signals are shown in Figure 12(a). 𝑜1, 1 and 𝑜2, 2 rotational scanning is done in 𝑥-𝑧 plane, resolution of image
represent the reflected energy from head phantom with and is increased and the number of false points is decreased, as
without tumor, respectively. ETR and LTR stand for early shown in Figure 13(b). This improvement is accomplished
time response and late time response, respectively. When a because the pulse from antenna in this case is transmitted to
cancerous human head is illuminated by an incident EM a more exact 𝑧 direction. However, the reconstructed image
wave, skin and other head tissue layer contained in the still contains some noise in the form of false point due to
scattering wave involve the optical region, which means that the inability of radiation pattern of the antenna to cover
these reflections contribute to the ETR. The tumor reflection the whole round phantom. To overcome these difficulties, a
is involved with the resonance region and contributes to the combined rotational scanning is carried out where the data
LTR. Hence, skin and tumor responses can be extracted by from rotational scanning in 𝑥-𝑦 and 𝑥-𝑧 plane are combined
investigating the turn-on time of LTR which is defined as together. The reconstructed image for combined scanning
International Journal of Biomedical Imaging 7

0.02 0.1 0.02

0.1

Patch

Substrate

(a) (b) (c)

Figure 4: Schematic diagram of (a) patch antenna with EBG; (b) geometry of rectangular EBG; (c) geometry of circular EBG.

6 6
4
4

Real permittivity and permeability

Real permittivity and permeability

2
0 2
−2
0
−4
−6 −2
−8
−4
−10
−12 −6
−14
3 4 5 6 7 8 9 10 11 12 13 3 4 5 6 7 8 9 10 11 12 13
Frequency (GHz) Frequency (GHz)

Re () Re ()
Re () Re ()
(a) (b)
Figure 5: Real permittivity and permeability of (a) circular and (b) rectangular EBG structures.

Brain
CSF

Dura

Bone

Fat
Tumor Skin

Figure 6: Schematic representation of the designed spherical

cancerous head phantom.

mode is shown in Figure 13(c) where the resolution and

shape of tumor are fairly good with minimal possibility of Figure 7: Simulation setup for the considered brain imaging
false diagnosis. Therefore microwave imaging via combined method.
8 International Journal of Biomedical Imaging

0 0

Reflection coefficient
Reflection coefficient

−5 −10

−20

(dB)
(dB)

−10

−15 −30

−20 −40
3 4 5 6 7 8 9 10 3 4 5 6 7 8 9 10
Frequency (GHz) Frequency (GHz)

CST CST
HFSS HFSS
(a) (b)
0

Reflection coefficient
0
−10
Reflection coefficient

−10
−20

(dB)
−20 −30
(dB)

−30 −40
−40 −50
−50 3 4 5 6 7 8 9 10
3 4 5 6 7 8 9 10 Frequency (GHz)
Frequency (GHz)
Rectangular
CST Circular
HFSS Normal
(c) (d)
Figure 8: Reflection coefficients of the patch antenna (a) without EBG, (b) with rectangular EBG, (c) and with circular EBG and (d)
comparison between three antennas.

7 is defined as a measure of SAR [25]. The mathematical

expression of SAR is as follows [25]:

6 𝜎 2 𝐽2
SAR = |𝐸| = , (5)
𝜌 𝜌𝜎
Gain (dBi)

5 where 𝐸 is the rms value of the electric field strength in

the tissue, 𝐽 is the current density, 𝜎 is the conductivity of
the head tissue, and 𝜌 is the density of head tissues. The
4 unit of SAR is usually represented as watt per kilogram.
Whenever human body is exposed to EM radiation, the
induced SAR inside the body tissue can be assessed using
3 different averaging techniques. The factors affecting SAR
5 5.5 6 6.5 7 7.5 8 calculation are the operating frequency, type of antenna, and
Frequency (GHz) distance between the scanning antenna and exposed surface
Circular
of human body. The international standard limit to SAR
Rectangular varies according to national reporting, testing requirement,
Normal and the network band. The 1 g average and 10 g average SAR
Figure 9: Gain of the designed antennas as a function of frequency. limt are 1.6 W/kg and 2 W/kg, set by US and EU standard,
respectively [26]. The distribution of SAR induced inside the
six-layered head tissues is shown in Figure 14, which is a 3D
rotational scanning of a multilayer human head phantom can plot obtained by numerical method based EM simulation.
reliably detect the presence of tumor inside the human head. To calculate the mass averaged SAR, the power loss density
is integrated over a cube with a defined mass of 1 g or 10 g
3.3. Evaluation of Specific Absorption Rate (SAR). As a brief and then the integral power loss is divided by the cube’s
exposure to radiation may cause severe health hazard, the mass [9]. The 1 g average and 10 g average SAR distributions
safety regulation of the envisaged brain imaging system must obtained by such numerical calculations are displayed in
be maintained by proper assessment of SAR. The deposition Figures 14(a) and 14(b), respectively. The 3D maximum SAR
of electromagnetic energy over time into human body tissue values are 0.922 W/kg and 0.695 W/kg, as shown in Figures
International Journal of Biomedical Imaging 9

Patch antenna with Patch antenna with

Normal patch antenna rectangular EBG Normal patch antenna rectangular EBG
0 0 0 0
30 10 30 30 10 30 30 10 30 30 10 30
0 0 5 5
0
60 −10 60 60 −10 60 60 60 60 0 60
−5
−20 −20 −10 −5
90 −30 90 90 −30 90 90 −15 90 90 −10 90

120 120 120 120 120 120 120 120

150 150 150 150 150 150 150 150

180 180 180 180

CST CST CST CST

HFSS HFSS HFSS HFSS

Patch antenna with Comparison between Patch antenna with Comparison between
circular EBG three antennas circular EBG three antennas
0 0 0 0
30 10 30 30 10 30 30 10 30 30 10 30
0 0 5 5
−10 0 0
60 60 60 −10 60 60 60 60 60
−20 −5 −5
−30 −20 −10 −10
90 −40 90 90 −30 90 90 −15 90 90 −15 90

120 120 120 120 120 120 120 120

150 150 150 150 150 150 150 150

180 180 180 180

CST Rectangular CST Rectangular

HFSS Circular HFSS Circular
Normal Normal
(a) (b)
Figure 10: Radiation patterns of the designed antennas at 7.3 GHz in (a) 𝐸-plane and (b) 𝐻-plane.

(a) (b)

(c) (d)
Figure 11: Position of antenna for scanning the phantom in (a) 𝑥-𝑦 plane, (b) reverse 𝑥-𝑦 plane, (c) 𝑥-𝑧 plane, and (d) reverse 𝑥-𝑧 plane.
10 International Journal of Biomedical Imaging

Time signals
0.75389

0.7

0.65
Time signals
0.8 0.6
0.6 Skin response Tumor
0.4 LTR response
0.55
0.2
0
−0.2 0.5
−0.4
−0.6 ETR
−0.8 0.42131
0 0.5 1 1.5 2 2.5 3 0.51973 0.55 0.6 0.65 0.7 0.75 0.7738
Time (ns) Time (ns)
o1, 1
o2, 2
(a) (b)
Time signals
0.28372
0.2
0.1
0
−0.1
−0.2
−0.3

−0.46269
1.517 1.6 1.7 1.8 1.9 2 2.1 2.2 2.3 2.4 2.4965
Time (ns)
(c)
Figure 12: (a) Reflected transient signal by simulating antenna using head phantom with and without tumor, (b) enlarged skin response, and
(c) enlarged tumor response.

0.15
5 10
0.08 False point
10 Tumor 0.1
Samples

0.06 20
Samples

15
20 Tumor
0.04 30
0.05
25 False points
0.02 40
30
0
100 200 300 400 500 600 700 800 900 1000 100 200 300 400 500 600 700 800 900 1000
Number of points Number of points
(a) (b)
0.15
10
20 Tumor
30 0.1
Samples

40
50
0.05
60
70
80 0
100 200 300 400 500 600 700 800 900 1000
Number of points
(c)

Figure 13: Reconstructed microwave image of tumor by scanning head phantom in (a) 𝑥-𝑦 plane, (b) 𝑥-𝑧 plane, and (c) 𝑥-𝑦 and 𝑥-𝑧 plane.
International Journal of Biomedical Imaging 11

(W/kg) (W/kg)
0.922 0.695
0.838 0.632
0.754 0.568
0.67 0.505
0.587 0.442
0.503 0.379
0.419 0.316
0.335 0.253
0.251 0.189
0.168 0.126
0.0838 0.0632
0 0
(a) (b)

Figure 14: (a) 1 g average and (b) 10 g average SAR induced inside the human head tissue.

14(a) and 14(b), respectively. It is clear from this figure References

that 1 g and 10 g average SAR values are well below the
maximum standard SAR limit, which ensures the safety of [1] R. L. Siegel, K. D. Miller, and A. Jemal, “Cancer statistics, 2017,”
CA: A Cancer Journal for Clinicians, vol. 67, no. 1, pp. 7–30, 2017.
our considered microwave brain imaging system. It can also
be noted from Figure 13 that the SAR values are higher in [2] B. J. Mohammed, A. M. Abbosh, S. Mustafa, and D. Ireland,
“Microwave system for head imaging,” IEEE Transactions on
the area of phantom surface near the antenna and quickly
Instrumentation and Measurement, vol. 63, no. 1, pp. 117–123,
decreases to zero in the distant areas of head phantom surface. 2014.
Therefore, no health risk regarding SAR is likely to occur
[3] M. Ostadrahimi, P. Mojabi, S. Noghanian, L. Shafai, S. Pistorius,
for the envisaged monostatic radar-based microwave brain and J. Lovetri, “A novel microwave tomography system based
imaging system which makes it a potential and reliable tool on the scattering probe technique,” IEEE Transactions on
to detect human brain tumor. Instrumentation and Measurement, vol. 61, no. 2, pp. 379–390,
2012.
4. Conclusion [4] F. S. G. B. Barnes, Handbook of Biological Effects of Electromag-
netic Fields, CRC/Taylor & Francis, Boca Raton, Fl, USA, 2007.
A brain imaging system working in microwave frequency range [5] N. K. Nikolova, “Microwave imaging for breast cancer,” IEEE
for human brain tumor detection is envisaged. An efficient Microwave Magazine, vol. 12, no. 7, pp. 78–94, 2011.
microstrip patch antenna is devised whose performance is [6] A. E. Souvorov, A. E. Bulyshev, S. Y. Semenov, R. H. Svenson,
improved by slotting circular EBG structure on the antenna and G. P. Tatsis, “Two-dimensional computer analysis of a
ground plane. The unique band gap feature of circular microwave flat antenna array for breast cancer tomography,”
EBG configuration boosts antenna performance parameters, IEEE Transactions on Microwave Theory and Techniques, vol. 48,
namely, reflection coefficient, gain, and bandwidth by con- no. 8, pp. 1413–1415, 2000.
siderable suppression of surface wave introduced on antenna [7] M. T. Islam, M. M. Islam, M. Samsuzzaman, M. R. Iqbal
substrate. The validation of the design is carried out by Faruque, and N. Misran, “A negative index metamaterial-
simulating the antennas with another software HFSS and inspired UWB antenna with an integration of complementary
comparing the results. The optimized and enhanced patch SRR and CLS unit cells for microwave imaging sensor applica-
antenna based on circular EBG is applied in the considered tions,” Sensors, vol. 15, no. 5, pp. 11601–11627, 2015.
imaging application. The antenna is used to scan a spherical [8] A. T. Mobashsher, K. S. Bialkowski, A. M. Abbosh, and S.
six-layer human head phantom model. The 𝑆-parameter Crozier, “Design and experimental evaluation of a non-invasive
microwave head imaging system for intracranial haemorrhage
data obtained from a combined rotational scanning mode
detection,” PLoS ONE, vol. 11, no. 4, Article ID e0152351, 2016.
performed in 𝑥-𝑦 and 𝑥-𝑧 plane is utilized in a confocal
[9] R. Inum, M. M. Rana, and M. A. Quader, “Modeling of an
microwave imaging algorithm to detect tumor inside the efficient microstrip patch antenna for microwave brain imaging
human brain. Finally, assessment of specific absorption rate system,” in Proceedings of the 3rd International Conference
(SAR) induced inside the human head phantom is carefully on Electrical Engineering and Information and Communication
done. The overall results obtained confirm that the studied Technology, iCEEiCT 2016, Dhaka, Bangladesh, September 2016.
brain imaging system can successfully diagnose brain tumor [10] S. A. Rezaeieh, A. Zamani, and A. M. Abbosh, “3-D Wideband
at an early stage while maintaining safety regulation of the Antenna for Head-Imaging System with Performance Verifica-
patient under test. tion in Brain Tumor Detection,” IEEE Antennas and Wireless
Propagation Letters, vol. 14, pp. 910–914, 2015.
Conflicts of Interest [11] M. Rokunuzzaman, M. Samsuzzaman, and M. T. Islam, “Uni-
directional Wideband 3-D Antenna for Human Head-Imaging
The authors declare that there are no conflicts of interest Application,” IEEE Antennas and Wireless Propagation Letters,
regarding the publication of this paper. vol. 16, pp. 169–172, 2017.
12 International Journal of Biomedical Imaging

[12] I. Singh and V. S. Tripathi, “Microstrip patch antenna and

its applications: a survey,” International Journal of Computer
Applications in Technology, vol. 2, no. 5, pp. 1595–1599, 2011.
[13] L. Zhang, Numerical characterization of electromagnetic band-
gap materials in printed antennas and arrays, [Ph.D. Disserta-
tion], Electrical Engineering Dept., University of California, Los
Angeles, Calif, USA, 2000.
[14] J. Liang and H. Y. D. Yang, “Microstrip patch antennas on tun-
able electromagnetic band-gap substrates,” IEEE Transactions
on Antennas and Propagation, vol. 57, no. 6, pp. 1612–1617, 2009.
[15] M. F. Abedin, M. Z. Azad, and M. Ali, “Wideband smaller
unit-cell planar EBG structures and their application,” IEEE
Transactions on Antennas and Propagation, vol. 56, no. 3, pp.
903–908, 2008.
[16] M. Mashevich, “Exposure of human peripheral blood lympho-
cytes to electromagnetic fields associated with cellular phones
leads to chromosomal instability,” Bioelectromagnetics (USA),
vol. 24, no. 82, 2003.
[17] A. T. Mobashsher and A. M. Abbosh, “Artificial human phan-
toms: Human proxy in testing microwave apparatuses that
have electromagnetic interaction with the human body,” IEEE
Microwave Magazine, vol. 16, no. 6, pp. 42–62, 2015.
[18] H. Zhang, A. O. El-Rayis, N. Haridas et al., “A smart antenna
array for brain cancer detection,” in Proceedings of the Lough-
borough Antennas & Propagation Conference, pp. 1–4, Lough-
borough, UK, 2011.
[19] S. Mustafa, A. M. Abbosh, and P. T. Nguyen, “Modeling human
head tissues using fourth-order debye model in convolution-
based three-dimensional finite-difference time-domain,” IEEE
Transactions on Antennas and Propagation, vol. 62, no. 3, pp.
1354–1361, 2014.
[20] C. A. Balanis, Antenna Theory Analysis And Design, John Wiley
& Sons, Inc., 3rd edition, 2005.
[21] O. Luukkonen, S. I. Maslovski, and S. A. Tretyakov, “A step-
wise Nicolson-Ross-Weir-based material parameter extraction
method,” IEEE Antennas and Wireless Propagation Letters, vol.
10, pp. 1295–1298, 2011.
[22] A. S. Bhadouria and M. Kumar, “Microstrip patch antenna for
radiolocation using DGS with improved gain and bandwidth,”
in Proceedings of the 2014 International Conference on Advances
in Engineering and Technology Research, ICAETR 2014, pp. 1–5,
India, August 2014.
[23] S. E. Mendhe and Y. P. Kosta, “Gain enhancement and
broadband using helical resonating metamaterial superstrate
in stacked microstrip patch antenna,” Microwave and Optical
Technology Letters, vol. 56, no. 9, pp. 1978–1982, 2014.
[24] S. K. Hong, W. S. Wall, T. D. Andreadis, and W. A. Davis, “Prac-
tical Implications of Pole Series Convergence and the Early-
time in Transient Backscatter,” NRL Memorandum Report,
Defense Technical Information Center, 2012.
[25] M. K. Hosain, A. Z. Kouzani, S. J. Tye et al., “Development of a
compact rectenna for wireless powering of a head-mountable
deep brain stimulation device,” IEEE Journal of Translational
Engineering in Health and Medicine, vol. 2, 2014.
[26] Antenna-Theory.com, Specific Absorption Rate (SAR), 2014,
http://www.antenna-theory.com/definitions/sar.php.
 International Journal of

Rotating Advances in
Machinery Multimedia

The Scientific
Engineering
Journal of
Journal of

Hindawi
World Journal
Hindawi Publishing Corporation Hindawi
Sensors
Hindawi Hindawi
www.hindawi.com Volume 2018 http://www.hindawi.com
www.hindawi.com Volume 2018
2013 www.hindawi.com Volume 2018 www.hindawi.com Volume 2018 www.hindawi.com Volume 2018

Journal of

Control Science
and Engineering

Advances in
Civil Engineering
Hindawi Hindawi
www.hindawi.com Volume 2018 www.hindawi.com Volume 2018

Submit your manuscripts at

www.hindawi.com

Journal of
Journal of Electrical and Computer
Robotics
Hindawi
Engineering
Hindawi
www.hindawi.com Volume 2018 www.hindawi.com Volume 2018

VLSI Design
Advances in
OptoElectronics
International Journal of

International Journal of
Modelling &
Simulation
Aerospace
Hindawi Volume 2018
Navigation and
Observation
Hindawi
www.hindawi.com Volume 2018
in Engineering
Hindawi
www.hindawi.com Volume 2018
Engineering
Hindawi
www.hindawi.com Volume 2018
Hindawi
www.hindawi.com www.hindawi.com Volume 2018

International Journal of
International Journal of Antennas and Active and Passive Advances in
Chemical Engineering Propagation Electronic Components Shock and Vibration Acoustics and Vibration
Hindawi Hindawi Hindawi Hindawi Hindawi
www.hindawi.com Volume 2018 www.hindawi.com Volume 2018 www.hindawi.com Volume 2018 www.hindawi.com Volume 2018 www.hindawi.com Volume 2018