Abstract
A critical review is given, of range and depth not previously available, of the dielectric dispersion mechanisms of biological materials from audio to microwave frequencies with specific regard to their medical significance and implications. Several hypotheses and models are proposed to correlate the biophysics of normal and pathological tissues with their respective dielectric properties. For the alpha-dispersion, Hypothesis 1 stresses the role of Ca[2+] ions to provide a mechanism unifying the cell membrane charge density, adhesiveness and contact inhibition changes induced by malignancy; Hypothesis 2 accounts for these characteristics through considerations of the role of the cell membrane electric field. The Hypotheses are not refuted nor able to be separated by the limited available experimental data. For the alpha-dispersion, quantitative predictions are made about the nature of its changes following the malignant transformation of the rat liver. The alpha-dispersion is synthesised for the normal rat liver using published morphological information. For malignant liver tissue, Model 1 examines mainly the consequences of increases in intracellular and static tissue conductivities; Model 2 further considers the effects of chromosomal proliferation and cell membrane outgrowths. The differences between the modelled malignant and normal liver permittivity peak at a frequency of about 2 MHz. In contrast, the differences in conductivity are relatively constant as a function of frequency, but are marginally greatest for Model 2 between 60-100 MHz. Equations are derived that indicate for normal and Model 1 malignant liver conductivity the presence of cross-over frequencies in the LF range. Several features of the predictions are in accord with published normal liver and hepatoma data. The analyses presented may have implications for tissue characterisation and electromagnetic hyperthermia therapy. The first use of a 6-port reflectometer for non-destructive RF and MW dielectric measurements is reported. Dielectric reference materials are used to examine the assumptions and limitations of the measurement technique -extensions are shown necessary and provided. Complex permittivities of a range of polar and biological materials are achieved. Those of several alcohols are generally consistent with recent theories concerning their molecular structure. The rat liver is measured in-vitro in states both of normality and pathophysiology (induced by phenobarbital). The measurements allowed comparison to predictions about the dielectric consequences of endoplasmic reticulum proliferation. The dielectric sensor realised in-vivo dielectric measurements of human skin and is shown sensitive to underlying tissues. Non-invasive dielectric-dependent imaging techniques are examined theoretically and experimentally. Both electrical impedance and microwave tomography have limitations that may render clinical applications impracticable. Applied potential tomography may offer a more feasible clinical technique for diagnostic imaging based on tissue conductivity differences.