IEEE MTT-V044-I10B (1996-10)

223 Pages·1996·26.585 MB·English

by  FANTOMASPING

IEEE T R A N S A C T I 0 N S O N

MICROWAVE THEORY
AND TECHNIQUES
A PUBLICATION OF THE IEEE MICROWAVE THEORY AND TECHNIQUES SOCIETY

OCTOBER 1996 VOLUME 44 NUMBER 10B IETMAB (ISSN 0018-9480)

[email protected]

PART II OF TWO PARTS

PAPERS
Editorial : Introduction to the Special Issue on Medical Application and Biological Effects of Micro - A. Rosen ; A.V. Vorst 1753 - 1754
Temperature control and thermal dosimetry by microwave radiometry in hyperthermia -
L. Dubois ; J.-P. Sozanski ; V. Tessier ; J.C. Camart ; J.-J. Fabre ; J. Pribetich ; M. Chive 1755 - 1761
Microwave prostatic hyperthermia: interest of urethral and rectal applicators combination-theoretical study and animal experimental results -
D. Despretz ; J.-C. Camart ; C. Michel ; J.-J. Fabre ; B. Prevost ; J.-P. Sozanski ; M. Chive 1762 - 1768
FDTD verification of deep-set brain tumor hyperthermia using a spherical microwave source distribution -
D. Dunn ; C.M. Rappaport ; A.J. Terzuoli 1769 - 1777
Use of the field-iteration method in studying the three-dimensional phased array for electromagnetic hyperthermia - Tianquan Deng 1778 - 1787
Analysis of focusing of pulse modulated microwave signals inside a tissue medium - K.S. Nikita ; N.K. Uzunogu 1788 - 1798
Focusing and impedance properties of conformable phased array antennas for microwave hyperthermia - R.M. Najafabadi ; A.F. Peterson 1799 - 1802
Development of ferrite core applicator system for deep-induction hyperthermia - Y. Kotsuka ; E. Hankui ; Y. Shigematsu 1803 - 1810
Modeling of various kinds of applicators used for microwave hyperthermia based on the FDTD method -
J.-C. Camart ; D. Despretz ; M. Chive ; J. Pribetich 1811 - 1818
A helical microwave antenna for welding plaque during balloon angioplasty - Ping Liu ; C.M. Rappaport 1819 - 1831
Monopole antennas for microwave catheter ablation - S. Labonte ; A. Blais ; S.R. Legault ; H.O. Ali ; L. Roy 1832 - 1840
A method for the in vitro testing of cardiac ablation catheters –
S.S. Hsu ; L. Hoh ; R.M. Rosenbaum ; A. Rosen ; P. Walinsky ; A.J. Greenspon 1841 - 1847
A finite element model of a microwave catheter for cardiac ablation - Z. Kaouk ; A. Khebir ; P. Savard 1848 - 1854
A study of the handset antenna and human body interaction - M. Okoniewski ; M.A. Stuchly 1855 - 1864
The dependence of EM energy absorption upon human head modeling at 900 MHz –
V. Hombach ; K. Meier ; M. Burkhardt ; E. Kuhn ; N. Kuster 1865 - 1873
Characteristics of the SAR distributions in a head exposed to electromagnetic fields radiated by a hand-held portable radio -
S.-I. Watanabe ; H. Taki ; T. Nojima ; O. Fujiwara 1874 - 1883
Electromagnetic absorption in the human head and neck for mobile telephones at 835 and 1900 MHz –
O.P. Gandhi ; G. Lazzi ; C.M. Furse 1884 - 1897
1990-1995 advances in investigating the interaction of microwave fields with the nervous system - A. Vander Vorst ; F. Duhamel 1898 - 1909
A model-driven approach to microwave diagnostics in biomedical applications - S. Caorsi ; G.L. Gragnani ; M. Pastorino ; M. Rebagliati 1910 - 1920
Image reconstruction of a complex cylinder illuminated by TE waves - Chien-Ching Chiu ; Po-Tsun Liu 1921 - 1927
A comparative study of four open-ended coaxial probe models for permittivity measurements of lossy dielectric/biological materials at microwave
frequencies - D. Berube ; F.M. Ghannouchi ; P. Savard 1928 - 1934

( Continued on back cover)
Evaluation of pulsed microwave influence on isolated hearts - M. Abbate ; G. Tine ; L. Zanforlin 1935 - 1941
No nonthermal effect observed under microwave irradiation of spinal cord - Jian Teng ; D.C. de Tournai ; F. Duhamel ; A. Vander Vorst 1942 - 1948
Soft and dry phantom modeling material using silicone rubber with carbon fiber - Y. Nikawa ; M. Chino ; K. Kikuchi 1949 - 1953
Broadband calibration of E-field probes in lossy media [mobile telephone safety application] -
K. Meier ; M. Burkhardt ; T. Schmid ; N. Kuster 1954 - 1962
Induced EM field in a layered eccentric spheres model of the head: plane-wave and localized source exposure
N.C. Skaropoulos ; M.P. Ioannidou ; D.P. Chrissoulidis 1963 - 1973

(end)
lEEE TRANSACTIONS ON MICROWAVE THECRY AND TECHNIQUES, VOL. 44, NO. IO, OCTOBER 1996 1753
Editorial :
Introduction to the Special Issue
on Medical Application and
Biological Effects of RF/Microwaves
I
N RECENT years, there has been a dramatic increase in the microwave antennas, and better catheter/probes at RF (kHz)
utilization of microwave/radio frel pency (RF) technologies frequencies. A number of papers dealing with cardiac therapy
in the commercial world, specific,illy in communications. are presented, specifically in the areas of cardiac ablation and
This has resulted in the widespread availability of improved microwave balloon angioplasty.
microwave/RF components and systems of much smaller sizes The third group deals with electromagnetic energy absorp-
and at much lower prices than were previously possible. This tion in human subjects, specifically in the human head and
has, in turn, resulted in a proliferaticn of many new ideas for neck. To assure coverage in this area as requested by many
the use of electromagnetic energy in medical therapies which of our MTT-S colleagues, the paper entitled “A Study of the
were not previously considered practical. Handset Antenna and Human Body Interaction,” by Michael
For example, the use of RF anti microwaves in cancer Okoniewski and Maria Stuchly was invited.
therapy in human subjects is well documented, and is currently The fourth section contains papers on modeling of a number
in use in many cancer centers. The utilization of RF in the of biological devices and systems, a model driven approach to
treatment of supraventricular arrhythinias in human subjects is microwave diagnostics, effects of microwave irradiation of the
currently employed by every major hospital. Similar modalities spinal cord, interaction of microwave fields with the nervous
are also used in human subjects for the treatment of benign system, an evaluation of the influence of pulsed microwave on
prostatic hyperplasia (BPH). isolated hearts, a study of open-ended coaxial probe models,
Despite these advances, there is 1:onsiderable effort being and image reconstruction of a complex cylinder illuminated
expended on improvement of the technology. Particularly, by TE waves.
the development of better antennas ;ind antenna systems (the It is obvious that the fields of biological effects of mi-
critical component in microwave th4:rapy) as well as that of crowave and microwave applications in medicine represent
better coupling mechanisms of RF mergy still dominate the strongly developing areas of research, which have great po-
area of hyperthermia. tential for stimulating new approaches in the future.
This special issue of the IEEE TRANSACTIONOSN Thanks are due to the reviewers for their thorough and
MKROWAVE THEORY and TECH^ IQUES presents current timely response, as well as to Dr. Hare1 D. Rosen for his
papers on the subjects of Biological Effects of Microwaves assistance in editing many of the accepted papers, and to
and Microwave Applications in Merlicine. It is divided into Danielle Rosen and Ronnie Kasian for their assistance in
four major sections. managing the project.
The first group of papers deals viith the broad subject of
RWmicrowave hyperthermia in canc<:rt reatment, and includes
subjects such as the utilization ol’ microwave radiometry ARVER OSEN,G uest Editor
in hyperthermia therapy, microwavc: prostatic hyperthermia, David Sarnoff Research Center
various RF/microwave focusing techniques and predictions, Princeton, New Jersey
techniques for achieving deep induct on hyperthermia, as well
as studies demonstrating microwave Iiyperthermia possibilities
in the treatment of tumors of the head. ANDREV ANDERV ORST, Guest Editor
The second group of papers deals with the optimization of Microwaves UCL
thermal effects in cardiology, in part through better design of Batiment Maxwell
Louvain-la-Neuve
Puhli\hcr Item Identifier Number S 0018-9~~80(96)07015-9. Belgium
0018-9480/96$05.00 0 I996 IEEE
1754 IEEE TRANSACTIONS ON MICROWAVE THEORY AND TECHNIQUES, VOL. 44, NO. IO, OCTOBER 1996
Arye Rosen (M’77-SM’80-F’92) was born in Israel in 1937. He received the B.S.E.E. degree
(cum laude) from Howard University, Washington, DC, the M.Sc.E. degree from Johns Hopkins
University, Baltimore, MD, the M.Sc. degree in physiology from Jefferson Medical College,
and the Ph.D. degree in electrical engineering from Drexel University, Philadelphia, PA.
He is a Senior Member of the Technical Staff at the David Sarnoff Research Center (formerly
RCA Laboratories) in Princeton, NJ, which he joined in 1967 and where he is currently
responsible for research and development in the areac of millimeter-wave devices and circuits
and microwave optical interaction. He currently holds an appointment as Center Professor in the
Center for Microwave and Light-Wave Engineering at Drexel University, where he has held an
appointment as Adjunct Profe\\or in the Department of Electrical and Computer Engineering
since 198 1. He also holds the title of Associate in Medicine at Jefferson Medical College,
where he hd? been engaged in research in the Division of Cardiology since 1970, specifically
in the areas of microwave balloon angioplasty and microwave/RF catheter ablation for the
treatment of cardiac arrhythmias, two subjects which are currently being investigated in several medical centers around the
world. The author of more than 150 technical papers, he is co-editor of High Power Optically Activated Solid-State Switches
(Boston: Artech House, 1993) and of New Frontiers in Medical Device Technology (New York: Wiley, 1995), and of six book
chapters in the field of engineering and medicine.
Dr. Rosen became a Fellow of the IEEE in 1992 “for innovation in semicondictor devices and circuits for use in microwave
systems and for microwave application\ to medicine.” He is a member of the IEEE MTT-S Technical Program Committee
since 1979, MTT-S Technical Committee Chairman on Biological Effects and Medical Applications, he has served as Associate
Editor of the IEEE JOURNALO F LIGHTWAVTE ECHNOLOG(JYL T), and IS a member of the Editorial Board of Microwave and
Optical Technology Letters. He has served on the Technical Committee for the IEEE International Conference on Microwaves
in Medicine held in Belgrade, Yugoslavia, in April 1991. He is also a Member-at-Large of the IEEE Health Care Engineering
Policy Committee, and has served as a Member of the IEEE Educational Actikities Board. He holds 45 United States patents
in the fields of engineering and medicine, and is the recipient of numerous achievement and professional awards, including a
1989 IEEE Region One Award “for significant contributions to microwave technology by the invention and development of
microwave balloon angioplasty.” His biography has been selected for inclusion in Murquis Who’s Who in the World 199551996,
AndrC Vander Vorst (M’64-SM’68-F’86) was born in Brussels, Belgium, in 1935. He
received the degrees of electrical and mechanical engineer in 19.58 and the Ph.D. degree
in applied sciences in 1965 from the Universit Catholique de Louvain, Belgium. In 1965, he
received the M.Sc. degree in electrical engineering from Massachusetts Institute of Technology,
Cambridge, MA. He is associated with the Universit Catholique de Louvain (UCL) where he
became an Assistant in 1958, Assistant Professor in 1962, Associate Professor in 1968, and
Professor in 1972.
From 1958 to 1964, he worked on fast switching of magnetic cores. With a NATO fellowship,
he was in the US. from 1964 to 1966, first at M.I.T., then at Stanford University, both in the
field of radio-astronomy. In 1966, he founded the Microwaves Laboratory at UCL, Belgium,
which he still heads, starting with research on loaded waveguides and cavities. The laboratory
is presently conducting research on atmospheric transmission and diffraction up to 300 GHz,
the methodology of designing and measuring activc and passive circuits up to 100 GHz,
and microwave bioelectromagnetics, namely the transmission between the peripheral and the central nervous system, using
microwave acupuncture as a stimulus. He was Head of the Electrical Engineering Department from 1970 to 1972, Dean of
Engineering from 1972 to 1975, Vice-president of the Academic Council of the university from 1973 to 1975, President of the
Open School in Economic ad Social Politics from 1973 to 1987, all at UCL, Belgium. He has authored or coauthored three
textbooks, several chapters, and a variety of scientific and technical papers in international journals and proceedings.
Dr. Vander Vorst is a member of the National Committee of URSI and of various committees on communications, microwaves,
and education. He has been active in IEEE Region 8 as well as in the European Microwave Conferences. He is a member of
Academia Europaea and The Electroinagnetics Academy. He has obtained the Sitel prize 1986, and the meritorious service
award of the Microwave Theory and Techniques Society, IEEE 1994. He became a Fellow of the IEEE in 1985 for his
contributions in atmospheric microwave propagation, satellite communication earth station design, and numerical analysis of
microwave components.
IEEE TRANSACTIONS ON MICROWAVE THEORY AND TECHNIQUES, VOL. 44, NO. 10, OCTOBER 1996 1755
Temperature Control and Thermal Dosimetry
by Microwave Radiometry in Hyperthermia
Luc Dubois, Jean-Pierre Sozanski, Virginie Tessier, Jean-Christophe Camart,
Jean-Jacques Fabre, Joseph Pribetich, and Maurice Chiv6
Abstract-This paper presents a synthesis of works undertaken 50 ’I /
by the Hyperthermia Group of Lille (France) concerning the
utilization of the microwave radiometry for the temperature
;
control in hyperthermia therapy. This technique of noninvasive 45
temperature control within the biological tissues has been in- c5
tegrated on many hyperthermia systems now commercialized. 0
We describe the principle of a new radiometer as well as the 40
calculation of radiometric signals. They allow a noninvasive .I-O-
determination of thermal maps inside tissues during the hy- cL
perthermia treatments. Many comparisons between theory and .- 35
experiment have validated our models of thermal dosimetry U2
whose provide a quantitative guidance for the planning of hy-
perthermia treatments. 30 . .
30 35 40 45 50
I. INTRODUCTION
Water Temperature (“C)
H
UMAN tissues spontaneously emit electromagnetic rddi-
Fig. 1. Calibration curves of a classical radiometer obtained for diffcrent
ations of thermal origin which can be measured by a very loads replacing the applicator. 50 51 load (/I = 0.0): Load 1 (p = 0.25);
sensitive receiver called “a radiometer” [ 11 and 121. When this Load 2 (/I = 0.1).
measurement is carried out in the microwave frequency range,
it is possible to evaluate the tissues temperature. Microwave 50 Ohms reference 50 Ohms reference
radiometry is used to detect thermal anomalies inside the tem erature Tr 1
human body (for example, the detection of breast cancer) [I]
and 131, but also to evaluate, noninvasively, the temperature
distribution in biological tissues [4]-[6]. microwave continuous
modulator
Since 1980, the Hyperthermia Group of Lille (composed short-CircU~
of the “Circuits et Applicateurs” Research group of the
Dtpartement Hyperfrtquences et Semiconducteurs of IEMN,
detector
the Centre Anti-Cancer Oscar Lambret and the Unit 279 IN-
SERM) has developed many hyperthermia systems combining 7im
microwave radiometers 161-1 101. It has been demonstrated
that these techniques made it possible to monitor hyperthermia
Fig, 2. Structure of the new radiometer with two internal temperature
systems, as well as, to plot thermal maps in clinical context references.
in order to optimize forthcoming hyperthermia sessions.
11. PHYSICAL PRINCIPLES OF MICROWAVREA DIOMETRY If T > IOK, at a frequency f and for a bandwith of 1 Hz,
B(f)i s expressed by the Rayleigh-Jeans relation
Any dissipative body emits spontaneous electromagnetic
radiations of thermal origin. In the microwave domain, the 2 - f 2 . k B . T
thermal noise power emitted by the body is directly propor- B(f)= c2
tional to the temperature and can be obtained by integrating NA . T
the spectrum brightness B(f)( Le., energy radiated per unit of with
apparent surface and per unit of solid angle). k . ~ Boltzmann’s constant (1.38 J . K-’);
r speed of light (3 . 10’ m x s-’);
Manuscript received October 12, 1995; revised February 22, 1996.
T absolute temperature of the body (Kelvins).
L. Dubois, V. Tessier, J.-C. Camart, .I-J. Fabre, and J. Pribeticb are with
the IEMN-DHS UMR CNRS 9929, UniversitC des Sciences et Technologies
The temperature of a dissipative body can thus be de-
de Lille, 59652 Villeneuve D’ Ascq Cedex, France.
J.-P. Sozanski is with the INSERM U 279, 59019 Lille Cedex, France. termined by a measurement of the electromagnetic power
M. ChivC is with the IEMN-DHS UMR CNRS 9929, UniversitC des radiated in a given frequency bandwidth. This measurement
Sciences et Technologies de Lille, 59652 Villeneuve D’A scq Cedex, France.
is achieved by radiometric systems which use an antenna as
He is also with the INSERM U 279, 59019 Lille Cedex, France.
Publisher Item Identifier S 001 8-9480(96)07022-6. an electromagnetic power captor in the microwave region. Let
0018-9480/96$05.00 0 1996 IEEE
1756 IEEE TRANSACTIONS ON MlCROWAVh THEORY AND TECHNIQUtS, VOL 44, NO IO, OCTOBER 1996
f
microwave continuous
short-circult Applicator ’L c-a-lib-ra-ti-on unit I
Fig. 3. Structure of the new radiometer containing a calibration unit constituted by two external calibrated sources raised, respectively, to tempera-
ture TI and Ti
us consider an antenna put on a dissipative body raised to a A. Principle of the Ideal Radiometer with Two
uniform temperature T. The power collected by the antenna Internal Temperature References
in a bandwidth Af is given by Nyquist’s formula The structure of this radiometer is given on the Fig. 2. It
P = ( l - p ) . k n . T .A f contains two reference sources constituted by 50 61 coaxial
loads, raised, respectively, to temperature TT1 and Tr2. A
with
microwave switch allows to select one of these two internal
kLI Boltzmann’s constant; temperature references.
T absolute temperature of the body (Kelvins); When the switch is in state “a,” the continuous voltage at
P power reflection coefficient at the interface the amplifier output is as follows:
applicator-lossy media. 1) modulator in state 1:
v,,
= G . kB . Af ’ T,,, (1)
111. MICROWAVREA DIOMETRISCY STEMS
2) modulator in state 2:
FOR BIOMEDICAALP PLICATIONS
V,2=G.ks.Af .[(l-p).T,+p.T,,1]. (2)
With the first generation of radiometers [I] and [2] the
output signal S is proportional to the difference in temperature When the switch is in state “b,” we obtain
1) modulator in state 1:
(Tz Tr)
-
S G . k~ Af . (1 - p) (T,- T,) v,, = G . ks . af . T~~ (3)
2) modulator in state 2:
with
+
V22 = G . k . .~ A f . [(1- p) . T, p . Tr2]. (4)
P power reflection coefficient at the interface
From these relations, we deduce the expression of the
applicator-lossy material;
reflection coefficient p and the expression of the temperature
temperature to measure;
TI
T, reference temperature; T,
G gain of the chain. v12 - v22
p=
Vll - V2l
We note that the output signal S depends on the gain G T, = (VII v12) . Tr2 - (V21+ - v22) ‘ ‘G.I (5)
and on the reflection coefficient p. Moreover, a preliminary v11 - VI2 - v21 v22
calibration of the system radiometer-applicator is necessary The value of the temperature T, thus calculated is now inde-
to obtain the T, information. The calibration is carried out pendant of the amplifier gain G and of the reflection coefficient
by putting the applicator in contact with a liquid emissive p. The value Tz represents the radiometric temperature (called
medium (salt water or physiological serum) that simulates the Trad) of the lossy material. It is an “average temperature”
biological tissues, whose temperature T is made to vary. Then of the volume of material coupled to the applicator in the
radiometer bandwidth. In the particular case of uniform tem-
we get a calibration curve Srnd = ,f(T).T his calibration
procedure takes about twenty minutes. perature, we have Trad = T,.
The resolution on the determination of the reflection coeffi-
But the calibration curve depends on the coefficient p. The
cient and for the radiometer sensitivity are given, respectively,
Fig. I shows its influence when the applicator is replaced by
different microwave loads. Consequently, we have studied a by
new radiometer allowing to free from the reflection coefficient
p and from the gain G.
DUBOIS et al.: TEMPERATURE CONTROL AND THERMAL DOSIMETRY BY MICROWAVE RADIOMETRY 1151
TABLE I
DETERMINATIO(WNIT H THE NEW RADIOMETER) OF TH~.R EFLECTIOCNO EFFICIENOFT DIFFERENMT ICROWAVLEO ADS
REPLACINTGH E APPLICATORA p Is DETERMINEEXDP ERIMENTABLYL AY S TANDARDDE VIATIOCNA LCULATION
I p (maximum value) 1 0.024 0.282 0.138 I
p (average value) 0.021 0.280 ~~~ 0.134
Ap (experiment) 0.003 0.003 0.004
P (theory) 00 0.25
AP (theow) 0.0026 0.0028
frequency (GHz) I p (network analyser) p (radiometer)
p(theory)=0.25 p(theory)=0.1 p(theory)=0.25 p(theory)=O.l
2.0 0.288 0.125
2.5 0.288 0.117
3.0 0.284 0.108
3.5 0.283 0.102
4.5 0.276 0.104
TABLE 111
CHARACTFRISOTFIC TSW O SETS OF 100 RADIOMETRMICE ASUREMPNTASc H1EVk.D WHEN THE APPI ICATOR IS PUT ON A THERMOSTATBEADT H
RAIWDTO 26 5OC AND THEN TO 45 Ioc AT7r id IC DET~RMINEEXDP ERIMENWLLY BY A STANDARD DEVIATIOCNAL CULATION
I I
Bath temperature measurements at 26.5OC measurements at 45.1OC
Trad (minimum value) I 26.52 I 45.17
1 Trad (maximum value) I 26.78 I 45.38 I
ATrad (exper~i~ ment) 0 058_ ______ ~ 0.0455
50 B. Calibration Procedure
F In practice, all elements of the radiometer present insertion
Y5 losses which modify the previous relations (1)-(4). We con-
45
sider consequently that the radiometer is ideal but with new
CFI
temperature references noted Trl, and Tradpe,du ced from two
E 40 external calibrated sources raised to temperature TI and T2.
I.-o- The block diagram of the radiometer is therefore slightly
cL modified in order to include a calibration unit (Fig. 3). The
;35 modulator is replaced by a switch with four positions and
the ways 3 and 4 are connected to the calibrated sources
(well-matched loads with same insertion losses) raised to
30
temperature TI and Tz.
30 35 40 45 50
To calibrate the radiometer, the four previous operations
Water Temperature ("C) (1)-(4) are done again but the applicator is now replaced
Fig. 4. Calibration curves of the new radiometer, obtained for different loads by the well-matched reference loads raised, respectively, to
replacing the applicator. 50 R load (p = 0.0); Load 1 (p = 0.23); Load 2 temperature TI and T2. The output voltages are
(p = 0.1).
1) With the unit calibration at temperature TI:
Vi1 =G . k~ . Af . T,1;
Viaa = G . k~ . A f . Ti
AT = fl . [ ( T + T g ) / d ni]s the theoretical sensitivity Vzl = G . k ~ .( A f ' Tr2;
of the Dicke radiometer [l] and [6] where TB and T are, v22=a G . krC . af . T~
respectively, the chain noise temperature and the time constant
of the synchronous detection. 2) With the unit calibration at temperature T2:
For medical applications the internal temperature references
T,1 and TT2 are raised, respectively, to 34°C and 55°C. Vi1 =G . Icg . Af . Trl;
1758 IEEE TRANSACTIONS ON MICROWAVE THEORY AND TECHNIQUES, VOL 44, NO IO, OCTOBER 1996
tadlomolflcs ignal
(norcmoanlllrzlebUu vllaolnu er)
0- -
0.0 1.0
I- 0.8.-0 9
0.7 0s
ground plane 2- 0.6 -- 0.7
0.5 0.6
9-
0.4 -a5
ceoxntdeurnctaot r catheter Water dielectric qoinndnuecr tot 4- a00 ..23 ~- 00..43
J--r 0.1 -0.2
0.0 .O.l
0 1a 2I 9I 41 5I 6I 8I b +Ocm
(a)
physiological Serum
3a6-343
30.8 - 92 a
290-308
27 2.200
-
23.4 27 2
plastic catheter 236 25.4
-with 6mm diameter 248.23~
200-21s
18.3 200
-
16.6 18 3
c
o i i i i i ii i b o ibcm
(b)
Fig. 6. Microwave radiometry can also be used in capacitive hyperthermia
W systems. (a) Normalized radiometric signal contribution of each subvolume
[calculated by (8)] and (b) thermal map obtained within an aquasonic gel after
(C) 50 min. heating by using a capacitive hyperthermia system (f = 13.56 MHz,
P,,,, = 93 W) controlled by microwave radiometry (1 .I GHz). The measured
Fig. 5. View of some applicators used on microwave hyperthermia systems and calculated radiometric temperatures are, respectively, 27OC and 27.7OC.
controlled by microwave radiometry. (a) Microstrip-microslot applicator for
external hyperthermia, (b) coaxial antenna for interstitial hyperthermia, and The calibration procedure takes about one min and it is
(c) endocavitary applicator for prostatic hyperthermia.
automatic without any modification of the position of the
applicator.
To verify and confirm this calibration procedure the ap-
plicator has been replaced by loads which present differents
values of reflection coefficient. These loads were plunged in a
thermostated bath which temperature T is made to vary.
The corresponding calibration curves are presented in
By combining these relations with the relation 5, we deduce
Fig. 4. The slope is equal to the unity and varies less than
the expression of the equivalent temperature references TT1
six per thousand when the reflection coefficient p varies from
and Tr2eas shown in (7a) at the bottom of the page. The
0.03-0.3.
radiometric temperature is therefore calculated by means of
We note therefore a straight improvement with regard to
relation 5 where the values TT1a nd Tr2 are replaced by TT1, results obtained with classical radiometers.
and Trze.
DUBOIS et al.: TEMPERATURE CONTROL AND THERMAL DOSIMETRY BY MICROWAVE RADIOMETRY 1759
eding line
PARAMETERS INPUT
n
-Incident Power
-Geometrical, dielectrical and thermal 0
characteristics of the media
-Geometrical parameters of the antenna 1
-Radiometric (Tr) and cutaneous (Tc) temperatures
2
Calculation of the absorbed power 3
4
5
from bioheat transfer equation
6
depth (cm)
(a)
aperture 50mm
I.
14 feeding line
water bolus (30°C)
0
YiQ 1
Thermal pattern display
2
Fig. 7. Flowchart of the thermal dosimetry software.
3
C. Performance
4
I) Measurement of the ReJlection Coeflcient: We have de-
termined, with this new radiometer (which operates in the
2-4 GHz frequency ranges), the reflection coefficient p of
different microwave loads replacing the applicator (Table I).
The values thus determined have been compared with the I I
network analyzer measurements achieved in the radiometer depth (cm)
bandwidth (for the same microwave loads). Results show a (b)
good agreement between the values measured by the network Fig. 8. (a) Thermal profile obtained along the feeding line of a mi-
analyzer and those determined with the radiometer (Table 11). crostrip-microdot applicator (diameter = 50 mm; E, = 4.9) laid on
a polyacrylamide gel, after 1 h heating (Pi,,, = 8.6 W; f = 915
2) Radiometric Temperature Measurements Sensitivity: The
MHz)-radiometric temperature Trnd (3 GHz) = 37.9OC; (b) calculated
sensitivity of the radiometer depends both on the reflection isotherms in the same conditions-calculated radiometric temperature T, (3
coefficient p and on the temperature T, to be measured. GHz) = 37.6OC.
From relation 7 it appears that the best sensitivity is obtained
when the temperature is situated between the internal reference
of radiometer is therefore completely in agreement with the
temperatures (TT1T, rz)a nd when the reflection coefficient is
theoretical analysis.
equal to zero.
Table I11 shows the characteristics of two sets of 100
Iv. EXPLOITATIOOFN R ADIOMETRISCI GNALS
radiometric measurements achieved when the applicator is put
on a thermostated bath (which stability temperature is better Radiometric signals received by a radiometer may be used
than 0.02"C) raised to 26.5"C and then to 45.1"C. for detecting thermal anomalies inside biological tissues [ 11
The delay for a radiometric measurement is around 5 s. and [3] or may be used for noninvasive temperature control
We note that the temperature resolution ATro*i s better than in hyperthermia treatments [6]-[lo]. In the case of microwave
0.05"C when the radiometric temperature (Trad) is equal to hyperthermia, the applicators (Fig. 5) are used both for heating
45.1"C. The value of ATTodi ncreases if Trad is not situated and for radiometric temperature measurements.
between T,.1 and T72. From these measurements and from radiometric signals
Experimental values of the temperature resolution are also in calculation it is also possible to evaluate noninvasively, the
good agreement with the theoretical values. So, the behavior temperature distribution inside the biological tissues [4]-[6].
1760 IEEE TRANSACTIONS ON MJCROWAVE THEORY AND TECHNIQUES, VOL. 44, NO. 10, OCTOBER 1996
A. Radiometric Signals Calculation aperture 50”
I4
The noise power P, measured through an applicator by
a radiometer centered on a frequency f~ (with a bandwidth
af)i s the integral summation of the elementary noise powers
emitted by each subvolume of the dissipative media and
multiplied by a weighting coefficient C. This coefficient
corresponds to the volume coupled to the applicator which
contributes to the noise power received by the radiometer. It
depends on the radiative diagram of the applicator at the fn
frequency and on the dielectric properties [I 11 and [ 121 of the
lossy media 1 1
P, = (1 - p) . dP (x:y ; z) . dx . dy . dz
with
dP (x,y , 2) = C(X;y , 2) . kg ‘ T(z,y , z) ’ af depth (em)
and Fig. 9. Comparison between calculated isotherms and temperatures mea-
sured by thermocouples on the axis of the applicator (diameter = 50 mm;
ici = 4.9) during a hyperthermia session on a patient. Experimental data:
f’,?),. = 21 W; f = 915 MHz; radiometric temperatures Trod (1 GHz) =
40. IOC, I-,.,, (3 GHz) = 38.6OC; temperatures measured by implemented
p is the power reflection coefficient in the input of the thermocouples: * 44.3OC; A43.0’C; 40.5OC; 638.6OC.
applicator; E is the electric field inside the lossy media when
thermocouples and bidimensional temperature profile recon-
the applicator is used in active mode at the frequency fn; and
struction.
is the electrical conductivity of media.
(T The computations are made on a desktop computer and
The correspond1ing radiometric temperature is given by
take a short CPU time (around 3 min.) which demonstrates
1
C(z, g. 2) . T(x:y : z) . dn: . dy . dz the possibility of simulation during the hyperthermia ses-
1 sion. However, in clinical situation, it’s necessary to know
Trad = 1
C(z, y, z) . d:r: . dy . dz accurately the structure of the heated tissues and to use multi-
frequency radiometry to improve the retrieval of temperature
Thus the radiometer detects an average of the temperature distributions.
distribution inside media, weighted by the squared electric field
V. CONCLUSION
pattern of the applicator used as a receiver. The electric field
E may be determined from many methods [7], [SI, [lo], and The new radiometer with two internal temperature refer-
[ 131-[ 151 and by applying the antenna reciprocity theorem. ences presents a great advantage as compared to the first
For example, we give in Fig. 6(a), the map of weighting generation one (with only one internal reference). Its calibra-
coefficients computed in the case of a radiometric antenna tion can be achieved very quickly (only one minute) using two
used in a hyperthermia capacitive system. calibrated sources. Another performance is that the radiometer
From this calculation we have determined the radiomet- measurement is independent of the reflection coefficient at the
ric temperature corresponding to the thermal map shown in applicator-tissues interface.
Fig. 6(b). The calculated and measured radiometric tempera- Microwave radiometry, used routinely since 1984, has
tures are in good agreement. proven its efficiency for noninvasive temperature control
during hyperthermia treatments. The radiometric signals
B. Application to Thermal Dosimetry
calculation allows to determine, noninvasively and with a
The radiometric signals calculation combined with the res- great accuracy, the thermal map within the tissues during
olution of bioheat transfer equation may be used to determine, hyperthermia sessions. So, our modeling allows to realize a
noninvasively, the thermal map inside tissues during hyper- thermal dosimetry and to provide a quantitative guidance for
thermia sessions [6j-[B], and [ lo]. the planning of hyperthermia treatments.
We give in Fig. 7 the flowchart of the thermal dosimetry
software [7] and [SI. In order to prove the validity of our REFERENCES
modeling, hyperthermia sessions on polyacrylamide gel were
[I] D. V. Land, “A clinical microwave thermography system,” IEE Proc.,
first performed. The Fig. 8 shows a great concordance between vol. 134, pt. A, no. 2, pp. 193-200, Feb. 1987.
theoretical and experimental isotherms when a microstrip- 121 M. ChivC, E. Constant, Y. Leroy, A. Mamouni, Y. Moschetto, D. D.
Nguyen, and J. P. Sozanski, “ProcCdC et dispositif de thermographie-
microslot applicator [7] and [16] is used for microwave
hyperthermie en microondes,” Brevet Franqais dCposC le 9 Janvier 1981,
heating. no. 8100682.
In the case of microwave hyperthermia on patients (Fig. 9), 131 B. Bocquet, J. C. Van de Velde, A. Mamouni, Y. Leroy, G. Giaux, J.
Delannoy. and D. Delvalee, “Microwave radiometric imaging at 3 GHz
many hyperthermia sessions have confirmed the good concor-
for the exploration of breast tumors,” IEEE Trans. Microwave Theory
dance between intratumoral temperatures measured by inserted Tech., vol. 38, pp. 791-793, 1990.