207
Interaction of Extremely Low–Frequency Electromagnetic Fields
of the models that are used in experimental and numerical dosimetries in the ELF electric field. The frequently used axisymmetric human body model represents the features
of the body, although not perfectly.
After the 1990s, exceedingly realistic and heterogeneous models of the human body
with small volume, block, and voxel (volumetric pixel) were developed as MRI technology emerged. Subsequently, MRI and CT scanners allowed researchers to create human
models at millimeter-resolution level (Dawson, de Moerloose, and Stuchly 1996; Dawson,
Moerloose, and Stuchly 1997a; Dawson, Potter, and Stuchly 2001b; Dawson 1997; Gandhi
and Chen 1992; Gandhi 1995). These newly developed models were posed in virtually any
position. These human models separate body parts into small volume, block, and voxel
of different conductivities. Heterogeneous models consisted of over 30 distinct human
organs and tissues. The size of the voxel ranged from 1 to 10 mm on a side. The conductivities of various organs were assigned the measured values. Data on dielectric properties
of tissues (conductivity and relative permittivity) for use in electromagnetic dosimetry
were established by measurement by Gabriel, Gabriel, and Corthout (1996a) and Gabriel,
Lau, and Gabriel (1996b,c). The conductivity data of organs and tissues at frequencies
below 1 MHz have been characterized by Gabriel (2005) and Gabriel, Peyman, and
Grand (2009). Gabriel, Lau, and Gabriel (1996b) stated that “it is possible that the dielectric parameters below 1 kHz may be under-corrected. This source of errors may affect the
permittivity values below 100 Hz by up to a factor of two or three.” Despite this issue,
the dielectric parameters of the biological tissues in the ELF region are generally used in
the calculation. It is well known that the individual variability of the dielectric parameters of biological tissues at low frequencies is greater than those at higher frequencies.
The human models for ELF magnetic field exposures have been developed at several
universities and organizations. Table 4.1 gives the characteristics of the different MRIbased male and female human models, and Figure 4.5 shows the visualized realistic
whole-body male and female human models obtained from the four different organizations. Historically, the University of Utah (USA) (Gandhi and Chen 1992), the University
of Victoria (Canada) (Xi, Stuchly, and Gandhi 1994a; Xi and Stuchly 1994b; Dawson, de
Moerloose, and Stuchly 1996; Dawson and Stuchly 1996), and the National Radiological
Protection Board (NRPB; now, HPA) (UK) (Dimbylow 1998) have developed human
models. The Brooks Air Force Base (USA) has developed a male rhesus monkey model and
used it for specific absorption rate (SAR) dosimetry calculations (Mason et al. 1999). The
National Institute of Information and Communication Technology (NICT, Japan) and
the Foundation for Research on Information Technologies in Society (IT’IS, Switzerland)
have also developed MRI-based and computer-aided-design (CAD)–based human models
(Table 4.1). The University of Victoria developed a phantom model based on MRI scan
data (Dawson, Caputa, and Stuchly 1997c). This phantom model was developed from a
head and torso model by the Yale Medical School (Zubal et al. 1994). It consisted of 30
organs and tissues, and cubic voxels of 3.6 × 3.6 × 3.6 mm 3 . This model was 1.77 m in height
and weighed 76 kg. Concerning heterogeneous human models, the University of Utah
developed a phantom model based on MRI scan data (Furse and Gandhi 1998; Gandhi
and Chen 1992). Gandhi and Chen (1992) created a whole-body model at a 2 × 2 × 3 mm 3
resolution based on MRI scans of an adult male subject who was 1.76 m tall and weighed
64 kg. The model was scaled up to 71 kg and segmented into 30 different tissue types.