International Journal of Bioelectromagnetism
Vol. 12, No. 1, pp. 7 - 11, 2010
www.ijbem.org
Body EMF Absorption: A Design Issue for
Implantable Medical Electronics
Qiang Fang
School of Electrical and Computer Engineering, RMIT University, Melbourne, Australia
Correspondence: Q. Fang, RMIT University, City Campus, VIC 3001, Australia
E-mail: [email protected], phone +61 3 99252432, fax +61 3 99252007
Abstract. Recently the human body implantable medical devices have been experiencing a rapid
development. Medical implants are an integration of sensing technologies, nano/micro technologies,
and wireless communication technologies and are developed as innovative approaches for in vivo body
health monitoring, essential life function supporting and diseases controlling/regulating. Those
miniaturized medical electronic systems have significantly improved the quality of life of many
patients. The majority of those human body implantable biomedical devices have wireless
communication capabilities. With the fast expansion of clinical adoption of those technologies, the
biological effects from the implantable electromagnetic field (EMF) sources have become a serious
concern. Though many researches have been conducted for Radio Frequency (RF) bioeffects from
external sources such as a mobile phone, few efforts have been spent on studying the effects from
implanted, the internal sources. This paper presents a review on various wireless communication
technologies and frequency bands used by implantable medical systems as well as the current adopted
EMF absorption safety guidelines and suggests that those safety guidelines are not sufficiently
addressed for implantable medical electronics design purposes.
Keywords: EMF absorptioin, SAR, Implantable electronics, MICS, Body Sensor Netoworks, Human Body Dielectric Model
1. Introduction
Recently many implantable medical devices have been developed for various medical applications
and have tremendously improved the quality of life (QoL) of many patients for whom the more
conventional and conservative treatment approaches are failed. Those implanted prosthesis are either
playing a crucial life supporting role such as a cardiac pace maker does, or are alleviating the
inconvenience in daily life caused by traumatic illnesses, such as stroke.
In principle, the medical implants can be classified into diagnostic and therapeutic two categories.
Within the first category, the implants for physiological status monitoring are the most popular ones.
Those implants are inserted into the patients’ body for in vivo monitoring of various physiological
signals such as Electrocardiograph (ECG), SvO2, blood pressure, body temperature, and blood glucose
level [Kjellström et al., 2004; Joseph et al., 2008; Klueha et al., 2005]. Also belong to this category are
the miniaturized medical image capturing devices such as the capsule endoscope. The therapeutic
implantable devices are actively involved in treatment procedures. The medical implants falling into
this category including surgical micro-robotics, various neuromuscular microstimulation systems,
insulin pump [Zhang et al., 2004], and other microelectromechanical systems (MEMS) based drug
delivery mechanism. Some implantable devices, e.g., microstimulators, can be used in a variety of
therapeutic applications to achieve different treatment goals such as to restore the control of paralyzed
limbs, enable bladder and bowel muscle control, maintain regular heart rhythm, and restore impaired
vision. Thanks for the fast advancement of microelectronics and MEMS technologies, the range of
medical devices and systems designed to be implanted into the human body is increasing rapidly.
In many cases, medical implants need data telemetry capability to exchange commands and data
with either other implanted devices or an external control unit. For some implantable devices, data
telemetry is also used for power transferring purpose. So far, a wide range of radio frequency bands
have been adopted by various proprietary implantable medical device manufacturers. Those frequency
bands range from medium-frequency (MF), high-frequency (HF), vey high frequency (VHF), and
ultrahigh frequency (UHF). The human body is a heterogeneous structure inside which the radio
frequency (RF) transmission undergoes a rather complex pathway due to multiple boundary reflection,
7
scattering, absorption and even refraction. So, different body parts have different absorption rates. The
body EMF absorption effect must be taken into account while designing the implantable medical
electronics because if it affects the system power consumption, the data transmission range, the
antenna design, and other key technical aspects. More importantly the body EMF absorption effect is a
potential hazard to the human body which has been often overlooked.
Although many international and national authority bodies have proposed safety guidelines to limit
the body EMF exposure, those guidelines are normally enacted for the EMF sources external to human
body. Most of those guidelines are concerning about the EMF thermal effects while the athermal
efforts are largely disregarded. The influence of electromagnetic emission from the in vivo sources,
such as those implanted electronic devices, has not been sufficiently investigated. This paper reviews
wireless communication technologies and frequency bands used by various implantable medical
systems and also reviews the popularly adopted EMF absorption safety guidelines.
2. Data Telemetry for Medical Implants
The two most popularly used frequency bands for implantable medical systems are Medical
Implant Communication Service (MICS) bands and Industrial, Scientific, Medical (ISM) bands. The
MICS bands (range of 402-405MHz) are the ultra low power radio frequency bands allocated by FCC
(Federal Communications Commission) for medical implant communication services while the ISM
bans are assigned by International Telecommunication Union [ITU Website] for industrial, scientific
and medical applications. The centre frequencies of ISM bands range from 6.78 MHz to 245 GHz. The
MICS bands are mainly used by human and animal implantable devices while the ISM bands are used
widely for a variety of wireless applications. The popularly adopted IEEE 802.15.1 (Bluetooth) and
IEEE 802.15.4 (ZigBee) are within the ISM bands.
The majority of implantable medical devices are equipped with wireless communication
capabilities for data transmission, command sending or power transportation. Both inductive coupling
and RF communication are employed for data telemetry.
2.1. Near Field Resonant Inductive Coupling
The basic principle of inductive coupling is the mutual induction of two closely positioned coils.
Though a traditional approach, the resonant inductive coupling is a popularly adopted wireless
communication technique for sending external commands to implants or retrieving acquired and stored
data out of the body. This technique has been used by various implants that don’t require high data
transmission rate such as implantable microstimulator, and implantable RFID transponder [Lee and
Lee, 2005; Lu et al., 2007]. Another popular usage of inductive coupling technique is the
transcutaneous wireless power transfer for implanted system [Shiba et al., 2008a, Shiba et al., 2008b].
Some designs have power telemetry and data telemetry combined in one system using a same carrier
frequency band to minimize the circuitry while other designs separate power and data carriers into
different frequency bands to reduce the interference and increase the data transmission rate. Most
transcutaneous telemetry systems use carrier frequency within either MF (medium frequency) or HF
(high frequency) bands, such as 13.56 MHz [Lu at al., 2007], 20 MHz, and 2 MHz [Lee and Lee,
2005]. For example, Lee et al designed an inductive charging system using a Class-E amplifier with
ASK modulations and a carrier frequency of 2 MHz to supply sufficient power for a low coupling
implantable microstimulator [Lee et al., 2004, Lee and Lee, 2005]. Due to the relative low efficient,
resonant inductive coupling is generally used by near field applications.
Because of the lower frequencies used by inductive coupling systems, inductive loop coils are used
to replace antennas, which sometimes make the systems bulky. Another disadvantage for this approach
is the low data transmission rate that it can achieve. In order to establish a coupling communication
session, the external coil and the internal coil need to be placed in a close proximity, i.e., in a near field
range, however, a precise positioning could be difficult to achieve because of the irregularity of the
human body.
2.2. Wireless Communication
The wireless based communication is essential for many active implantable medical systems,
which need to create a communication session from inside of a body. The ISM bands have been
adopted by a variety of medical implants as carrier frequencies for data telemetry. For example, in
[Valdastri P et al., 2008], an implantable data telemetry system using 2.4 GHz ZigBee was designed
and implemented. An early design of implantable heart rate and blood oxygen sensor continuously
monitoring venous oxygen saturation (SvO2) uses 27 MHz pulse modulated radio signal to transmit
8
the acquired signals to an AM radio receiver outside the body [Holmstrom et al., 1998]. Comparing
ISM bands, the MICS bands are much more popularly used by many active implantable medical
systems, such as glucose monitor, blood pressure monitor and bionic eyes, as they are specific
proposed for implantable applications. MICS band has a maximum bandwidth of 300 kHz and a
typical coverage range of 2 meters [FCC Website]. In order to ensure the safety, the maximum power
is limited to 25 μW Equivalent Radiated Power (ERP) in air. The ERP is defined as the transmitted
signal power measured on external surface of human body, i.e, the skin in many cases.
3. Safety Guidelines for Body EMF Absorption and
3.1. The body EMF exposure safety guidelines
Due to the widespread use of electrical appliances and mobile telecommunication devices, the
human exposure to EMF has become omnipresent, especially in metropolitan areas. Many national and
international organizations have enacted safety guidelines based on recognized scientific research
outcomes. The IEEE C95.1-2005 [IEEE 2005] Standard specifies the safety levels of human RF
exposure from 2 KHz to 300 GHz. The International Commission on Non-Ionizing Radiation
Protection (ICNIRP) also published its guidelines for limiting exposure to time-varying EMF up to 300
GHz [ICNIRP 1998]. The appropriate physical parameters used to specify the exposure restrictions are
frequency dependent. Given the fact that most implantable medical systems have a wireless
communication frequency between 100 KHz to 10GHz, the specific absorption rate (SAR) and current
density (for frequency up to 10 MHz) are generally followed.
ICNIRP guidelines define basic restriction as well as reference levels for general public exposure
and occupational exposure respectively. For a frequency range from 100KHz to 10GHz, the basic
exposure restrictions recommended by ICNIRP for general public are 0.08 W/kg for whole body
average SAR and 2 W/kg for localized SAR (head and trunk) while the basic restrictions for
occupational exposures are 0.4 W/kg for whole average SAR and 10 W/kg for localized SAR. For
frequency from 100KHz to 10 MHz, restrictions on current density are set to f/100 mA m-2 for
occupational exposure and f/500 mA m-2 for general public exposure [ICNIRP 1998].
From 2005, the IEEE C95.1 standard has been unified with ICNIRP guidelines with respect to
SAR limits in the frequency range from 100 kHz to 3 GHz. The IEEE C95.1-2005 also uses the
Maximum Permissible Exposure (MPE) to restrict the external fields and induced current quantities
such as electrical field strength, magnetic field strength and power density for both controlled and
uncontrolled environments. The MPE restrictions can be derived from SAR restrictions. Based on
IEEE C95.1-2005, the human exposures to implanted medical systems are considered to be partial
body exposures in an uncontrolled environment. IEEE also provides detailed recommendations for
numerical computation of spatial-averaged specific absorption rate [IEEE 2002]. While designing the
medical implants, the occupational exposure restriction (10 W/kg for localized SAR) is generally
followed by considering a medical implant as a part of the medical procedures controlled by medical
professionals [Chirwa et al., 2003].
Many implantable system designers follow either the IEEE C95.1 or ICNIRP guidelines to address
the EMF exposure safety issues, though those guidelines are not explicitly designed for regulating the
EMF exposure from implantable electronics devices. Those guidelines are drafted based on extensive
published scientific research outcomes. Those published researches include epidemiological studies,
residential cancer studies, laboratory studies, and cellular and animal studies [ICNIRP 1998; IEEE
2005]. The majority of those research employ only external EMF sources. Though some investigations
on in vitro and in vivo effects have been performed, few experiments were done using internal sources.
Moreover, those guidelines mainly consider the thermal effects of EMF, i.e., the increase of tissue
temperature due to the EMF exposure. The reported athermal interactions were not taken into account
during the guideline formation due to the prevalent opinion that there are no sufficient and consistent
evidences to support the biological significance of athermal effects [IEEE 2005]. Nevertheless, it is
worth to point out that medical implants reside inside human body for a very long term (could be many
years) and have a close spatial proximity to many key organs. The current researches on EMF
interaction with biomacromolecules such as protein and DNA are far from sufficient. Thus, particular
caution shall be taken to design implantable medical systems using those published guidelines. Rather
than following the occupational exposure standards, it is suggested here to use the more restricted
general public exposure limits (2 W/kg for localized SAR) to design the medical implants to minimize
the possible adverse effects because the EMF exposure from an internal implanted source is continuous
9
for many years. However, such a measure will impose great challenges to biomedical engineers
designing the implantable systems.
3.2 EMF exposure experiment from internal sources and human body modeling
It is commonly known that the human body absorbs energy from electromagnetic waves generated
by any transmission circuits. Thus, electrical properties of human tissue have to be considered when
addressing any implantable telemetry module. The human body has a complicated shape and is of
heterogeneous nature due to a variety of tissues and organs which have different permittivity,
conductivity and magnetic permeability. Different types of body tissue have different electrical
properties. For instance, at 150MHz the blood has the conductivity of 1.644S and the relative
permittivity of 65.2 while the conductivity for fat is only 0.035S and the relative permittivity is 5.79
[Chirwa et al., 2003]. With the rapid advancement of wireless communication technologies, many
research efforts have been spent on studying human body EMF interaction with various
electromagnetic fields and radiators [Scanlon and Evans, 1997; Scanlon and Evans, 2001; Lazzi and
Gandhi, 1998; Scanlon and Evans, 2000; Toftgard et al., 1993; Jensen and Rahmat-Samii, 1995;
Okoniewski and Stuchly, 1996; Colburn and Rahmat-Samii, 1998; Iddan et al., 2000]. Due to the
limited condition imposed on human body and the difficulty to measure reference levels in vivo, few
studies have been carried out on electromagnetic fields from implanted sources. In the meantime, it is
of great importance to use computational human body models to investigate the body EMF interactions
as an indirect approach.
There are three basic human body dielectric models, namely the simple homogenous model, the
semi-segmented model, and the full-segmented model. The homogeneous body model is the most
popularly used one due to its computational simplicity [Han and Hall, 2008; Toftgard et al., 1993]. In
this model, the surrounding and adjacent tissues have homogeneous dielectric properties and the body
shape can been determined by using animation software. In many practical cases, those models are
used together to form a composite body model to reflect the actual situation.
In [Scanlon and Evans, 2000] RF absorption of a 418MHz radio telemeter semi-implanted in the
human vagina is compared with simulation results through human EMF absorption modeling. A
composite human model was used in this investigation with three semi-segmented models within
100mm, 200mm and 300mm varying tissue depths from the source respectively and one fully
homogeneous 5mm model (5 mm depth of skin) constructed as three layers (skin, fat and muscle).
Based on the assumption that the source surrounded tissues are responsible for most of the bulk
absorption loss, a full segment human body model was not used in this study. The comparison results
find that the radiation pattern and field strength do not vary much in different segment depths and
hence prove the originally assumption, i.e., the surrounding tissues absorb the most of the EMF energy,
is correct.
Another detailed investigation on the electromagnetic radiation inside human body (inside
intestine) is done by [Chirwa et al., 2003].The investigated frequency range is between 150MHz and
1.2GHz. Compared to the method utilized in [Scanlon and Evans, 2000], the human modelling in this
investigation was also constructed using the finite-difference time-domain (FDTD) method based on
the data from the Visible Human Project but simulated using full-segmented body model. Both near
field and far field performances are taken into consideration. It is found that both the near and far-field
results show that maximum radiation is experienced on the anterior side of the body from radiation
sources in the small intestine. The results also suggest that implanted devices are most efficient in the
450 to 900 MHz range. It needs to be noted that this experiment was carried out using EMF sources
placed inside intestine, thus the results might not be applicable to subcutaneous implants.
4. Conclusion
In order to optimally design the implantable electronic medical devices, the radiation
characteristics of implanted EMF sources need to be fully understood. In the meantime, in order to
ensure the safety guards set by international originations such as ICNIRP, the SAR limits also need to
be followed. However, the radiation characteristics of sources inside the body cannot be readily
calculated due to the complexity of the human body and the experimental researches on this area are
also far from comprehensive. The current safety guidelines only consider the short term immediately
biological effect, such as the elevation of tissue temperature, while the long term possible adverse
effects due to either thermal or athermal effects, such as the increased risk of cancer, are not taken into
account. In addition, those guidelines are predominately based on the investigations using external
EMF sources. Thus, it is suggested that more researches on body EMF absorption from internal EMF
10
sources need to be done and the a conservative measure of SAR limit set by ICNIRP needs to be taken
during the implant electronics design process for the time being.
References
Chirwa LC, Hammond PA, Roy S, Cumming DRS. Electromagnetic radiation from ingested sources in the human intestine
between 150MHz and 1.2GHz. IEEE Transactions on Biomedical Engineering, 50(4):484-492, 2003.
Colburn JS, Rahmat-Samii Y. Human proximity effects on circular polarized handset antennas in personal satellite
communications. IEEE Transactions on Antennas and Propagation, 46(6): 813–820, 1998.
FCC Website: http://wireless.fcc.gov/
Guillory KS, Misener AK, Pungor A. Hybrid RF/IR transcutaneous telemetry for power and high-bandwidth data. In Proceedings
of 26th Annual International Conference of the IEEE Engineering in Medicine and Biology Society, 2004, 4338 – 4340.
Hao Y and Hall PS. On-Body Antennas and Propagation: Recent Development. IEICE Transactions on Communication. E91–
B(6):1682-1688, 2008
ICNIRP 1998 Guidelines for limiting exposure to time-varying electric, magnetic, and electromagnetic fields (up to 300 GHz)
Health Phys. 74 494–522, 1998.
Iddan G, Meron G, Glukhovsky A, Swain P. Wireless capsule endoscopy. Nature, 405(6785):417, 2000.
IEEE 2002 Recommended Practice for Measurements and Computations of Radio Frequency Electromagnetic Fields With
Respect to Human Exposure to such Fields, 100 kHz to 300 GHz, IEEE Standard C95.3-2002, 2002.
IEEE 2005 c95.1-2005 IEEE Standard for Safety Levels with Respect to Human Exposure to Radio Frequency Electromagnetic
Fields, 3 kHz to 300 GHz
ITU Website: http://www.itu.int/net/home/index.aspx
Jensen MA, Rahmat-Samii Y. The electromagnetic interaction of handset antennas and a human in personal communications.
Proceedings of the IEEE, 83(1): 7–17, 1985.
Joseph P, Gaetano B, Bonato P. Implantable Electronics. IEEE Pervasive Computing, 7(1): 12–13, 2008.
Kjellström B, Linde C, Bennett T, Ohlsson A. Six years follow-up of an implanted SvO2 sensor in the right ventricle. European
Journal of Heart Failure, 6(5): 627-634, 2004.
Klueha U, Dorskya DI, Kreutzera DL. Enhancement of implantable glucose sensor function in vivo using gene transfer-induced
neovascularization. Biomaterials, 26(10): 1155–1163, 2005.
Lazzi G. Gandhi OP. On modeling and personal dosimetry of cellular telephone helical antennas with FDTD code. IEEE
Transactions on Antennas Propagation. 46(4): 525–530, 1998.
Lee SY, Lee SC, Chen JJ. VLSI Implementation of Implantable Wireless Power and Data Transmission Micro-Stimulator for
Neuromuscular Stimulation. IEICE Transactions on Electronics, 87(6): 1062-1068, 2004.
Lee SY, Lee SC. An Implantable Wireless Bidirectional Communication Microstimulator for Neuromuscular Stimulation. IEEE
Transactions on Circuits and Systems – I: Regular Papers, 52(12): 2526-2538, 2005.
Lu HM, Goldsmith C, Cauller L, Lee JB. MEMS-Based Inductively Coupled RFID Transponder for Implantable Wireless Sensor
Applications. IEEE Transactions on Magnetics, 43(6):2412-2414 ,2007.
Nils Holmstrom N, Nilsson P, Carlsten J, Bowald S. Long-term in vivo experience of an electrochemical sensor using the
potential step technique for measurement of mixed venous oxygen pressure. Biosensors & Bioelectronics, 13(12): 1287–1295,
1998.
Okoniewski M, Stuchly MA. A study of the handset antenna and human body interaction. IEEE Transactions on Microwave
Theory and Techniques, 44(10): 1855–1864, 1996.
Scanlon WG, Evans NE. RF performance of a 418-MHz radio telemeter packaged for human vaginal placement. IEEE
Transactions on Biomedical Engineering, 44(5): 427–430, 1997.
Scanlon WG, Evans NE. Radiowave propagation from a tissue implanted source at 418 MHz and 916.5 MHz. IEEE Transanction
on Biomedical Engineering, 47(4): 527–534, Apr. 2000.
Scanlon WG, Evans NE, “Numerical analysis of bodyworn UHF antenna systems,” IEE Electronics & Communication
Engineering Journal, 13(2): 53–64, 2001.
Shiba K, Nagato T, Tsuji T, Koshiji K. Energy Transmission Transformer for a Wireless Capsule Endoscope: Analysis of
Specific Absorption Rate and Current Density in Biological Tissue. IEEE Transactions on Biomedical Engineering, 55(7): 18641871, 2008a.
Shiba K, Nukaya M, Tsuji T, Koshiji K. Analysis of Current Density and Specific Absorption Rate in Biological Tissue
Surrounding Transcutaneous Transformer for an Artificial Heart. IEEE Transactions on Biomedical Engineering, 55(1): 205-213,
2008b.
Toftgard J, Hornsleth SN, Andersen J. Effects on portable antennas of the presence of a person. IEEE Transactions on Antennas
Propagation, 41(6): 739–746, 1993.
Valdastri P, Rossi S, Menciassi A, Lionetti V, Bernini F, Recchia FA, Dario P. An implantable ZigBee ready telemetric platform
for in vivo monitoring of physiological parameters. Sensors and Actuators A, 142(1): 369–378, 2008.
Zhang M, Tarn T, Xi N. Micro/nano-devices for controlled drug delivery. In the Proceedings of 2004 IEEE International
Conference on Robotics and Automation, 2004, 2068 – 2073.
11