272
Electromagnetic Fields in Biological Systems
5.2.2 Tissue Dielectric Properties
Human tissues contain insulating materials (lipids) and electrical charges (ions, electrically polarized molecules, etc.). Therefore, they can be viewed as a weakly conducting
medium (dielectric). The dielectric properties of human tissues are directly related to
the amount of RF energy that is absorbed and converted into heat, due to the increase
in their molecular translational and rotational kinetic energy. Exact knowledge of the
dielectric properties of biological tissues has become essential in RF dosimetry in order
to enable more accurate assessment of human exposure to RF EM fields emitted by
mobile communication devices.
The currently most comprehensive, complete, and best-known database of RF dielectric properties of body tissues is based on the work of Gabriel et al., which provides
dispersive parameter models for several important body tissues (Gabriel, Gabriel, and
Corthout 1996). Before this work was published, most data on the dielectric properties
of tissues used in numerical dosimetry were obtained from measurements of animals,
such as pigs, sheep, or rabbits. Based on these measurements, Gabriel et al. proposed a
parametric model using a Cole–Cole analysis for several body tissues in the frequency
range from 10 Hz to 20 GHz. This model has been used extensively in numerical dosimetry for humans and animals.
Several other research efforts have been performed to assess the dielectric properties
of human tissues. For example, Schmid et al. (2007). presented results for the dielectric properties of human gray matter tissue based on a sample of tissues. The sample
was larger than ever reported before with respect to the number of brains, number of
measurement locations, and freshness of the tissue. In the studies of Lazebnik et al.
(2007b), the results of a large-scale, multi-institutional study characterizing the dielectric properties of normal breast tissue samples obtained from reduction surgeries were
reported. Parameters for Debye models describing the dielectric properties of normal
and malignant breast tissues were derived in the studies of Lazebnik et al. (2007a).
Finally, O’Rourke et al. (2007) have recently characterized the dielectric properties of in
vivo and ex vivo human liver tissues between 0.5 GHz and 20 GHz.
However, the electrical properties of human tissues are not known with a high
degree of precision and accurate knowledge of these properties is currently evolving. Almost all the work mentioned in Gabriel, Gabriel, and Corthout (1996) is based
either on measurements on excised animal tissues about 2 h after death or, to a very
small extent, on human autopsy material obtained more than 24 h after death. Only a
few publications have reported in vivo measurements on animal tissues (Stuchly and
Stuchly 1984; Surowiec et al. 1986). Furthermore, some data produced by other workers (Campbell and Land 1992) deviate significantly from the frequently used values in
Gabriel, Gabriel, and Corthout (1996). Gabriel et al. have discussed the uncertainties
in the measurement of dielectric properties and highlighted the major contribution
of random variations from repeat measurements (Gabriel and Peyman 2006). These
ambiguities can have important consequences on computed dosimetric quantities and
may result in significantly different SAR distributions inside the human body (Gandhi,
Lazzi, and Furse 1996; Hombach et al. 1996; Okoniewski and Stuchly 1996). For example, it appears that variations up to 50% in the convective coefficient between the head
Electromagnetic Fields in Biological Systems
5.2.2 Tissue Dielectric Properties
Human tissues contain insulating materials (lipids) and electrical charges (ions, electrically polarized molecules, etc.). Therefore, they can be viewed as a weakly conducting
medium (dielectric). The dielectric properties of human tissues are directly related to
the amount of RF energy that is absorbed and converted into heat, due to the increase
in their molecular translational and rotational kinetic energy. Exact knowledge of the
dielectric properties of biological tissues has become essential in RF dosimetry in order
to enable more accurate assessment of human exposure to RF EM fields emitted by
mobile communication devices.
The currently most comprehensive, complete, and best-known database of RF dielectric properties of body tissues is based on the work of Gabriel et al., which provides
dispersive parameter models for several important body tissues (Gabriel, Gabriel, and
Corthout 1996). Before this work was published, most data on the dielectric properties
of tissues used in numerical dosimetry were obtained from measurements of animals,
such as pigs, sheep, or rabbits. Based on these measurements, Gabriel et al. proposed a
parametric model using a Cole–Cole analysis for several body tissues in the frequency
range from 10 Hz to 20 GHz. This model has been used extensively in numerical dosimetry for humans and animals.
Several other research efforts have been performed to assess the dielectric properties
of human tissues. For example, Schmid et al. (2007). presented results for the dielectric properties of human gray matter tissue based on a sample of tissues. The sample
was larger than ever reported before with respect to the number of brains, number of
measurement locations, and freshness of the tissue. In the studies of Lazebnik et al.
(2007b), the results of a large-scale, multi-institutional study characterizing the dielectric properties of normal breast tissue samples obtained from reduction surgeries were
reported. Parameters for Debye models describing the dielectric properties of normal
and malignant breast tissues were derived in the studies of Lazebnik et al. (2007a).
Finally, O’Rourke et al. (2007) have recently characterized the dielectric properties of in
vivo and ex vivo human liver tissues between 0.5 GHz and 20 GHz.
However, the electrical properties of human tissues are not known with a high
degree of precision and accurate knowledge of these properties is currently evolving. Almost all the work mentioned in Gabriel, Gabriel, and Corthout (1996) is based
either on measurements on excised animal tissues about 2 h after death or, to a very
small extent, on human autopsy material obtained more than 24 h after death. Only a
few publications have reported in vivo measurements on animal tissues (Stuchly and
Stuchly 1984; Surowiec et al. 1986). Furthermore, some data produced by other workers (Campbell and Land 1992) deviate significantly from the frequently used values in
Gabriel, Gabriel, and Corthout (1996). Gabriel et al. have discussed the uncertainties
in the measurement of dielectric properties and highlighted the major contribution
of random variations from repeat measurements (Gabriel and Peyman 2006). These
ambiguities can have important consequences on computed dosimetric quantities and
may result in significantly different SAR distributions inside the human body (Gandhi,
Lazzi, and Furse 1996; Hombach et al. 1996; Okoniewski and Stuchly 1996). For example, it appears that variations up to 50% in the convective coefficient between the head