Search

This paper presents a measurement procedure for analyzing the dielectric properties of cells using descriptive statistics. The study focuses on four cancer cell lines (MDA-MB-231 and MCF-7 breast cancer, SaOS-2, and 143B osteosarcoma) and DMEM culture medium, utilizing the Lorentzian fit model of the return-loss function. The measurements are performed using a circular patch resonator with a 40 mm diameter, powered by a miniVNA operating in the frequency range of 1 MHz to 3 GHz. Eight specimens are prepared for each group to ensure reliability, and the return loss is recorded ten times for each specimen. Various statistical parameters are calculated and evaluated, including the average value, standard deviation, coefficient of variation, and relative error between the average and the first values. The results demonstrate that one single acquisition highly represents the entire set of ten data points, especially for the resonant frequency, with an accuracy error lower than 0.05%. These findings have significant implications for the methodological approach to detecting cells’ dielectric properties, as they substantially reduce time and preserve the specimens without compromising the accuracy of the experimental results.

The measurement of biological tissues’ dielectric properties plays a crucial role in determining the state of health, and recent studies have reported microwave biosensing to be an innovative method with great potential in this field. Research has been conducted from the tissue level to the cellular level but, to date, cellular adhesion has never been considered. In addition, conventional systems for diagnosing tumor aggressiveness, such as a biopsy, are rather expensive and invasive. Here, we propose a novel microwave approach for biosensing adherent cancer cells with different malignancy degrees. A circular patch resonator was designed adjusting its structure to a standard Petri dish and a network analyzer was employed. Then, the resonator was realized and used to test two groups of different cancer cell lines, based on various tumor types and aggressiveness: low- and high-aggressive osteosarcoma cell lines (SaOS-2 and 143B, respectively), and low- and high-aggressive breast cancer cell lines (MCF-7 and MDA-MB-231, respectively). The experimental results showed that the sensitivity of the sensor was high, in particular when measuring the resonant frequency. Finally, the sensor showed a good ability to distinguish low-metastatic and high-metastatic cells, paving the way to the development of more complex measurement systems for noninvasive tissue diagnosis.

In this work, the dielectric permittivity of four kinds of wood (Fir, Poplar, Oak, and Beech Tree), used in Italian Artworks and structures, was characterized at different humidity levels. Measurements were carried out using three different probes connected to a bench vector network analyzer: a standard WR90 X-band waveguide, a WR430 waveguide, and an open-ended coaxial probe. In particular, we investigated the dispersion model for the four wood species, showing how a log-fit model of the open-ended data presents a determination coefficient R2 > 0.990 in the 1–12 GHz frequency range. This result has proven helpful to fill the frequency gap between the measurements obtained at different water contents with the two waveguide probes showing an R2 > 0.93. Furthermore, correlating the log-fit vertical shift with the water content, it was possible to find a calibration curve with a linear characteristic. These experimental results will be helpful for on-site non-invasive water monitoring of wooden artworks or structures. Moreover, the final results show how the open-ended coaxial probe, with a measurement deviation lower than 7% from the waveguide measurements, may be used directly as a non-invasive sensor for on-site measurements.