ISEF 2003 11th International Symposium on Electromagnetic Fields in Electrical Engineering

Maribor, Slovenia, September 18-20, 2003

III-18. ELECTROMAGNETIC HYPERTHERMIA FOUNDATIONS

AND COMPUTER MODELLING
Jolanta Plewako, Andrzej Krawczyk, Barbara Grochowicz
Rzeszw University of Technology, Rzeszw, Poland, e-mail: jplewako@man.rzeszow.pl
Technical University of Czstochowa, Czstochowa, Poland, e-mail: krawczyk@iel.waw.pl
Technical University of Opole, Opole, Poland, e-mail: bgr@polo.po.opole.pl

Abstract The usage of heating power as a healing system has been well-known for long time. A
completely new motivation, however, came up when heat treating was recognized as a new and
promising form of cancer therapy. It was found that cancer growth was stopped at temperature higher
than about 42 C. It created new subject of research in the area of application of electromagnetic fields in
medicine. In the paper the two main methods of heating have been described and two kinds of devices
have been presented. Some advantages and drawbacks of the methods are discussed.

Introduction
Hyperthermia, a procedure in which body tissue is exposed to high temperatures (up to 42C and above),
is under investigation to assess its effectiveness in the treatment of cancer. Hyperthermia has by now become
the chosen treatment for some important physiotherapeutic pathologies of the muscle-tendon apparatus, and
plays a fundamental role integrating with other methodologies in the more general rehabilitating program.
Scientists believe that heat may help shrink tumors by damaging cells or depriving them of substances
they need to live. They are studying local, regional, and whole-body hyperthermia, using external and internal
heating devices. Hyperthermia is almost always used with other forms of therapy (radiation therapy,
chemotherapy, and biological therapy) in attempt to increase their effectiveness.

Main problems
The wave propagates from the surface of the tissues towards the inside, and while it proceeds it is adsorbed,
loosing electromagnetic energy that is transformed into heat. The mechanisms of heat deposition in tissues by
electromagnetic fields is followed. When the tissue's electric dipoles (both permanent and induced) oscillate in
response to the E-field of an applied wave, heat is generated by a process analogous to friction. Similarly, when
free charges (electrons and ions) in the tissue are set in motion by the E-field, collisions with immobile atoms
and molecules in the tissue generate heat. The propensity of the tissue to produce heat for a given sinusoidal Efield magnitude is determined by the values of the imaginary part of its relative permittivity " and its
conductivity . It is important that the internal E-field (i.e. the electric field inside the body) is responsible for
the heat generation. In addition, the internal H-field is not directly responsible for heating because tissue has a
permeability close to that of free space with no magnetic losses. But the time-varying H-field produces a
resulting internal E-field (eddy currents) and in this way it causes heating of tissue.
The human body has an intricate structure (roughly stratified structure of the muscle-skeletal apparatus:
skin, fat, muscle, bone), and that patients have a variety of physiological and psychological responses to
hyperthermia treatment. The underlying principle is that a patient's responses must be monitored and considered
in later designs, so as to reduce the patient's complaints and ultimately improve the efficiency by which the
treatment is delivered.
From the very beginning of the application in question, there have been two essential problems to
overcome:
Generation of heat within the region of interest leaving all the vicinity of it unaffected.
Monitoring and controlling the temperature, both in the region of interest and its vicinity.
The above problems are attempted to be solved by using different methods of heating, like capacitive,
inductive, by microwave radiation, or by ultrasounds. It seems, however, that the therapy is still at the stage of
medical research.
Sawomir Wiak, Andrzej Krawczyk, Mladen Trlep (eds.), Computer Engineering in Applied Electromagnetism, 337342.
2005 Springer. Printed in Great Britain.

338

Plewako et al.

Hyperthermia in Cancer Treatment

Hyperthermia activates the immune system. In normal tissues, blood vessels open up, (dilate) when heat
is applied, dissipating the heat and cooling down the cell environment. Unlike healthy cells, a tumor is a tightly
packed group of cells, and circulation is restricted and sluggish. When heat is applied to the tumor, vital nutrients
and oxygen are cut off from the tumor cells. Heat above 41C also pushes cancer cells toward acidosis
(decreased cellular pH) which decreases the cells viability and transplantability. This results in a collapse of the
tumor's vascular system and destruction of the cancer cells. Tumor masses tend to have hypoxic (oxygen
deprived) cells within the inner part of the tumor. These cells are resistant to radiation, but they are very sensitive
to heat. This is why, hyperthermia is an ideal companion to radiation: radiation kills the oxygenated outer cells,
while hyperthermia acts on the inner low-oxygen cells, oxygenating them which makes them more susceptible to
radiation damage. It is also thought that induced accumulation of proteins, induced by hyperthermia, inhibits the
malignant cells from repairing the damage sustained.
Techniques in clinical hyperthermia can be classified into three categories:
whole body,

regional,
local hyperthermia.
Whole-body heating is used to treat metastatic cancer1 that has spread throughout the body. It can be
accomplished using warm-water blankets, hot wax, inductive coils (like those in electric blankets), or thermal
chambers (similar to large incubators). In regional hyperthermia, an organ or a limb is heated. Magnets and
devices that produce high energy are placed over the region to be heated. In another approach, called perfusion2,
some of the patient's blood is removed, heated, and then pumped (perfused) into the region that is to be heated
internally. Local hyperthermia refers to the heat that is applied to a very small area, such as a tumor. The area
may be heated externally with high-frequency waves aimed at a tumor from a device outside the body. To
achieve internal heating, one of several types of sterile probes may be used, including thin, heated wires or
hollow tubes filled with warm water; implanted microwave antennae; and radio-frequency electrodes.
The two major categories of applicators are developed for electromagnetic hyperthermia:
noninvasive applicators, which use devices external to the body to produce the internal E-field:
capacitive,
inductive,
radiative,
invasive applicators, which penetrate the body either through the skin or in natural body orifices. The
invasive applicators are listed in order of the types of external fields that are principally responsible
for the internal E-field:
electrodes,
radiative antennas.

Capacitive Applicators
A capacitive applicator is composed of two conducting electrodes which are placed on or near the
surface of the body (Fig. 1). The electrodes can have various shapes and sizes. A voltage source is connected
across the electrodes, producing an E-field stretching throughout the volume between them. The E-field lines
terminate on charges contained in the electrodes. Since these applicators are often intended to heat deeper
tissues, the frequency of the voltage source is relatively low (in the high kHz to low MHz range).
The advantages of the capacitive-type applicator are based upon its simplicity. The placement and shape
of the electrodes can be tailored to the location of the region that is to be heated. It is relatively easy to visualize
the paths that the field lines take. Also, the electrodes can be curved to match the skin contour.

Cancer that has spread from the place in which it started to other parts of the body
Bathing an organ or tissue with a fluid. In regional perfusion, a specific area of the body (usually an arm or a leg) receives high doses of
anticancer drugs through a blood vessel. Such a procedure is performed to treat cancer that has not spread
2

III-18. Electromagnetic Hyperthermia Foundations and Computer Modelling

339

Fig. 1. Simple example of a capacitive applicator [1]

The drawbacks of this type of applicators are that the fields generated in the tissue are not optimum for
preferentially heating deep tumors because the E-fields are mostly perpendicular to the body surface, where there
are fat layers and if muscle or muscle-like, tissue is beneath the fat. The boundary conditions for normal E-field
components combined with the lower permittivity of fat means that the E-field in the fat is much higher than in
the muscle. Even though fat is less loss, the higher E- field results in higher energy deposition, often overheating
the fat layer. A common tendency with capacitive applicators is to burn areas on the surface of the body when
attempting to heat deeper tissue. The E-fields concentrates at the edges of metallic electrodes. Spots on the skin
are vulnerable to burns near the corners of the applicator. To reduce this problem by spreading out the fields,
water boluses (nonmetallic containers of water) are sometimes placed between the electrodes and the skin. The
water in the bolus can even be chilled and recirculated to provide some conductive cooling of the skin.

Inductive Applicators
In this type of applicator, an external coil or some other means of generating high currents near the body
is used to produce an H-field inside the body (Fig. 2). The magnetic field itself, according to the mechanism of
heating described above, does not produce any heat, but if the H-field is time-varying it will induce an internal
E-field for heating. These applicators are generally provided to deep heating, which suggests again lower
frequency. However, since the generation of the internal E-field is proportional to the time rate of change of the
H-field, the frequency should be high enough to produce a sufficient internal E-field. Operating frequencies are
generally in the low MHz range.

Fig. 2. Simple example of an inductive applicator for hyperthermia [1]

The H-field lines run longitudinally through the body, then encircle the coil outside the body because H-field
lines must close upon themselves. Depending on the exact geometry of the coil windings and the size of the
body, the density of the H-fields inside any cross section of the body can be fairly uniform. Other forms of coil
applicators can also be used, such as pancake coils or saddle-shaped coils, or specially shaped conductors that
bring currents to the surface of the body.
All inductive applicators share common advantages and disadvantages.
One advantage of this type of applicator is its relative insensitivity to the coupling conditions (since tissue is
nonmagnetic, the exact position of the body with respect to the coil does not affect the H-field pattern). It is
comfortable for patient because it allows him moving without any harm of treatment. In addition, the tuning of
the coil in the resonant electrical circuit of the source is forgiving of exact body positioning.

340

Plewako et al.

One disadvantage is that centrally located tumors would not be heated effectively. Heating is greatest at the
periphery, so surface heating is a major concern with inductive applicators, as it is with capacitive applicators.
Because the E-field lines are produced by the time-varying H-field they encircle the H-field lines. There is a
center of rotation for the E-field and here the field is zero. Also, eddy currents are zero in tissue. The field and
current grow linearly toward the periphery of the cylinder. Because power deposition P is proportional to the
square of the E-field, the heating pattern has a parabolic shape.
If the tissue properties are not uniform as in this simple example, eddy currents will not follow a
radially linear profile and will be more irregular. This sometimes can be used to advantage. For example, a highconductivity tumor surrounded by lower conductivity tissue will have a local eddy-current pattern flowing
around the approximate center of the tumor. The local eddy-current patterns can lead to increased heating of a
deep tumor, but the amount of improvement depends on the conductivities of the tissues involved, which may
vary considerably from case to case.

Radiative Applicators
This class of applicators relies upon the coupling of E and H to carry electromagnetic energy into the
tissue. They operate either at higher frequencies when localized surface heating is needed or at lower frequencies
when deeper penetration is desired. The applicator and feed configurations are chosen to maximize the coupling
of the launched wave into the tissues. One version of a radiative applicator is shown in Fig. 3. It basically
consists of an open-ended waveguide that is coupled to the skin with a quarter-wavelength matching slab. The
waveguide is loaded on both sides with dielectric strips. This produces a mode structure, thus giving a more
uniform pattern in the transverse direction than an unloaded waveguide. The size of the waveguide dictates its
relatively high operating frequency, namely 2450 MHz, so it is appropriate for heating superficial tumors.

Fig. 3. A radiative hyperthermia applicator consisting of an open-ended waveguide[1]

To make the applicator size more compact, microstrip radiators have also been developed. These
applicators are lightweight and can even be made flexible, so they are more convement to use than the larger,
heavier waveguides. They operate at higher frequencies (433 MHz to 2450 MHz), so they are meant for
localized superficial heating. All electromagnetic radiative applicators face the same tradeoff between depth of
penetration, applicator size, and localizing ability. Fig. 4 shows the penetration characteristics for planewaves of
various frequencies into a dielectric halfspace whose properties are similar to those of muscle. The higherfrequency waves are clearly attenuated quickly by the tissue due to their high loss. Although the waves coming
from practical applicators are not planewaves and the body certainly is not an infinite halfspace, this same
general behavior is expected to apply to radiative applicators. Note from Fig. 4 that to penetrate to reasonable
depths (say, beyond 7 or 8 cm), the frequency must be about 100 MHz or lower. The wavelength in muscle,
therefore, is quite large-about 30 cm.

III-18. Electromagnetic Hyperthermia Foundations and Computer Modelling

341

Fig. 4. The penetration of planewaves of various frequencies into a dielectric halfspace with the
properties of muscle [1]
A radiator is not very effective unless at least one dimension of the radiating structure is one-half of a
wavelength or larger. If the frequency is lower, the applicator will be even larger. This means it will be rather
bulky and heavy, and more important, the energy coming from the applicator will spread out due to diffraction.
Localized heating is difficult at the low frequencies that will penetrate deeply. They are drawbacks of this type of
applicators.
Ultrasound waves obey these same laws but with different constants and with a much different outcome.
Ultrasound's advantage of being able to penetrate deeply with small-wavelength beams is one reason ultrasound
energy is being seriously considered for hyperthermia therapy. A disadvantage, however, is that ultrasound will
not effectively penetrate bone or air, so treatment is limited to regions of the body where access is through soft
tissue.

Invasive Applicators
To circumvent the difficulty of obtaining deep, localized heating patterns from external electromagnetic
applicators, some investigators use invasive probes. These probes are placed in natural cavities of the body (if
the tumor is nearby) or directly through the skin. The cavity applicators are often designed as thin radiating
antennas. Higher frequencies are used to get good radiation efficiency from the small antennas, and penetration
depth is not as critical as with external applicators. The probes that pierce the skin may also be small radiating
antennas or may be an array of lower-frequency electrodes. In the latter case, conduction current in the tissues
produces the heating.
The advantage of invasive probes is that the heat can be localized with more precision and in a smaller
volume at depth than with external applicators. One disadvantage is much more uncomfortable for the patient.
Also, even using multiple probes does not assure uniform heating; there still may be considerable no uniformity
to the power deposition pattern depending on the placement and individual patterns from the probes.

Hyperthermia Simulation and Treatment Planning

As part of a bigger research project ( Sonderforschungsbereich "Hyperthermia: Scientific Methods and
Clinical Applications") new algorithms for simulating and planning regional hyperthermia are developed at ZIB.
These include methods for segmenting medical image data, generating tetrahedral patient models, solving
Maxwell's equations and the problem of heat transport, as well as novel visualization methods. All these methods
are integrated into a single, flexible, easy-to-use software system, called HyperPlan.
HyperPlan is able to simulate of the electric and thermal processes in the patient's body numerically. In
order to perform a complete simulation some intermediate steps have to be executed. For every step a set of
special-purpose modules is provided. All steps can be controlled by means of 3D visualization methods. In order
to simulate the electric and thermal processes in the patient's body two different mathematical equations are
solved in HyperPlan, Maxwell's equation in inhomogeneous media and the bio-heat transfer equation. This is
achieved by means of modern adaptive multi-level finite-element methods. These methods require the user to
create a tetrahedral model of the patient, describing the shape of the tumor and of other tissue compartments.
Tools to carry out this task are integrated into HyperPlan. As a result, electromagnetic fields and power
deposition in the patient's body are obtained. Using this information, a temperature distribution is calculated in a
subsequent step. Finally, an optimized setting of applicator control parameters is computed.

342

Plewako et al.

Fig. 5. System overview

Summary
It should be clear from the previous discussion that one area that remains problematic with electromagnetic
hyperthermia is the ability to heat deeply in a well-controlled and localized manner. Too often there is surface
overheating that accompanies deep heating, regardless of the type of applicator used. Based upon the concepts of
electromagnetics, it seems unlikely that this problem will be easily solved. When localized superficial heating is
desired, on the other hand, several of the approaches, in particular the small radiating applicators, are successful.
Other engineering issues remain. These include the need for multiple-point temperature measurements for
accurate and thorough monitoring. Treatment planning will require accurate characterization of the applicator
deposition pattern and the tissue parameters, as well as a numerical technique to predict the resultant heating
pattern. Tissue perfusion significantly modifies the temperature distribution for any given power deposition
pattern, often in a time-variable and unpredictable way. Still, the promise of even a partially successful therapy
for cancer spurs the continued study of hyperthermia.
References
[1] C.H. Durney, D.A.Christensen, The Basic Introduction to Bioelectromagnetics, Boca Raton: CRC Press,
2000
[2] W. Andra, H. Nowak, Magnetism in Medicine, Berlin:Wiley-VCH, 1999
[3] M.W. Dewhirst, T.L. Phillips, T.V. Samulski., RTOG quality assurance guidelines for clinical trials using
hyperthermia. Int. J. Radiation Oncology Biol. Phys., 1990; 18:1249-1259.
[4] M.J. Piket-May, A. Taflove, W.C. Lin, D.S. Katz, V. Sathiaseelan, B.B. Mittal, Initial results for
automated computational modeling of patient-specific electromagnetic hyperthermia, IEEE Trans.
Biomedical Engineering, 1992; 39: 226-237
[5] H.A.Vera, Q.J.E.Chong, S.L. Leija, M.Y.Hernndez, Electromagnetic hyperthermia: an adjuvant treatment
on cancerous cells, biological and physical principles, Rev Mex Ing Biomed 2001; 22 (2): 78-88
[6] T. Sugahara, I.Yamamoto, V.Ostapenko, How to develop hyperthermia equipment for deep-seated tumors,
http://www.taishitsu.or.jp/hyperthermia/hp1-e.html
[7] http://www.nebraskahealthsystem.com/cancer/cancer.cfm
[8] http://www.esho.info/professionals/hyperthermia/
[9] http://www.veramed.de/index2.html
[10] http://www.hot-oncotherm.de/english/prof/index.htm