Determining the Relative Permittivity of Deep Embedded Biological Tissues

Abstract

This letter introduces a novel in vivo method for determining the dielectric properties of deep human tissues, including lungs. A surgery-free method is proposed for an on-body monitoring system to evaluate the electrical properties of internal body organs and to detect irregularities in real time. The method uses a set of electrodes (16 or 40 MHz in operational frequency) to obtain data that relates the electrical properties of the underlying biological tissue(s) to the measured port scattering parameters. By using multiple electrodes, effects from outer layers (skin, fat, and muscle) can be suppressed, allowing for an accurate characterization of deeper layers. Specifically, the dielectric constant is expressed as a weighted sum of the scattering parameters from each probe. The weights of the scattering coefficients are then determined through a training process to ensure that various changes in the outer layers do not contaminate the extraction of the dielectric constant(s) associated with deeper tissues.Index Terms-Biomedical sensor, human health monitoring, signal-to-noise ratio (SNR), tissue dielectric properties, wearable electronics.

A Wearable Wrap-Around Sensor for Monitoring Deep Tissue Electric Properties

“…The feasibility of (3) was already validated using a smaller sensor of 16 ports in [7]. The extracted ε r , from [7], was found accurate to within a small error of < 8%. Next, we validate the wrap-around sensor with measurements.…”

“…As in [7], we choose to represent the lung dielectric constant ε r as a weighted sum of the transmission coefficients obtained at the passive ports. Among the available 65 ports we use the data from 12 electrodes and 11 ports (1 active and 10 passive ports; ports refer to voltage difference between adjacent electrodes, see Fig.…”

“…The process of calibrating out the outer layers of the body amounts to targeting or localizing expression (2). Specifically, as done in [7], S i j values are computationally collected for a variety of skin, fat, muscle, and bone (i.e. outer layers) electrical properties.…”

“…As compared to [7], our goal in this paper is to evaluate the wrap-around sensor and verify its capability to detect variation in ε r across the torso. By contrast, [7] only verified the expression (1) for a 16 electrodes sensor over a uniform cross-section. Here, the wrap-around sensor contains 65 electrodes and is partitioned into subgroups.…”

“…A specific application considered is that of detecting lung fluid and tumor. A simpler version of the proposed sensor was fabricated and tested in [7]. In this paper, we expand on this concept to develop a wrap-around version of the sensor with many more electrodes.…”