Microwave Resonant Sensor for Non-invasive Characterization of Biological Tissues
Résumé
Context: Microwave sensing appears to be an open wide field to investigate medical applications, such as monitoring of vital signs (temperature, arterial pressure, ...), following different kinds of pathologies (cancer, glucose level ...) and aid for medical diagnosis. It offers an alternative to determine the dielectric properties of biological tissues through the development of local non-invasive and/or embedded sensors,
giving thus a kind of imaging by the dielectric contrast. Moreover, RF communications links between several sensors can be developed to realize “Body Area Networks”.
Methods: Biological tissues having high dielectric permittivity and losses in the microwave frequency domain (around 1 GHz), a resonant dielectric characterization method is used to obtain a good sensitivity. The experimental setup is based on the measured changes of electrical characteristics of the resonator (resonance frequency and its shift and broadening) when a biological tissue is applied on it. In our
case, the sensor is a microstrip ring resonator operating in a two-port configuration at a fundamental frequency of 1 GHz. It consists of a meander loop in order to reduce its dimensions. Besides, an original excitation is proposed leading to small perturbation of the resonator when high dielectric losses material is characterized. This increased greatly the sensitivity of the method to obtain the dielectric properties of the
samples. Dielectric parameters are determined by fitting S parameters measurements results with those of simulations using electromagnetic software’s (HFSS, CST).
Results: Several biological tissues of animal origin were measured ex-vivo in the frequency range 0.5–5 GHz. The dielectric parameters obtained by this method are consistent with values proposed in databases or obtained by other researchers. A very good agreement between simulations and measurements is obtained
leading to a good extraction of permittivity and losses of the tissues.
Conclusions: This paper presents an improved microwave sensor, either for reduced dimensions as for sensitivity, able to perform dielectric characterization of material having high complex permittivity such as biological tissues. Experiments and electromagnetic simulations have been achieved on several animal tissues (chicken, beef, pork ...), and results are in good agreement with literature. Works are in progress to optimize this sensor as an applicator for medical applications.