Epidermal Devices for High Accuracy Measurement of Skin’s Bioimpedance

Abstract

Skin, the largest organ of human body, performs several crucial functions. To name a few, it prevents trans-epidermal water loss and protects against pathogens. Skin health is strongly related to its hydration level. Indeed, dehydration of skin can cause various diseases such as skin cancer, eczema, acne, itching, cracking of the stratum corneum, obesity, diabetes etc. Monitoring the skin hydration level is therefore important for both the dermatology and cosmetology. Due to the dependence on water content, skin’s bioimpedance is a strong reflection of its hydration level. Skin’s bioimpedance is proposed as a non-invasive and low-cost method for monitoring its hydration level. It is measured by injecting current into skin through electrodes and measuring the related voltage drop across skin. Since the dielectric and conductive properties of skin layers is highly dependent on frequency, therefore, the bioimpedance is usually characterized over a broad frequency range of 10 Hz to 1 MHz. Recently, skin-like epidermal electrodes are developed for bioimpedance characterization with minimum mechanical and thermal loading to the skin. Skin comprises epidermis and dermis layers that together are known as viable skin (VS). It has adipose tissue (AT) layer and muscle (M) beneath it and stratum corneum (SC), comprising dead cells, on top of it. There are several challenges to characterize the skin’s bioimpedance. Highly insulating stratum corneum prevents the impedance measurement of the deeper layers termed as deep bioimpedance. Indeed, conventional two wire impedance bioimpedance measurements are dominated by stratum corneum. Four-wires Kelvin measurement are widely adopted for low impedance measurements and could potentially exclude the contribution of stratum corneum to the skin’s bioimpedance. Deep bioimpedance measurement has a strong dependence on electrodes geometry as wider electrodes can in principle concentrate the current in deeper layers. Nonetheless, the electrodes introduce parasitic capacitances that can also cause errors in the measurements by providing the alternate current paths. This thesis presents the COMSOL Multiphysics based finite element method (FEM) simulations of skin’s bioimpedance. Firstly, the electric potential distribution, equipotential lines and current distribution across SC, VS, AT, and M is modelled. To quantify the current through each layer, the concept of current tubes is defined. For instance, the VS current tube is collection of the current streamlines that connect the electrodes and pass through the VS layer at the central, vertical line midway between the electrodes. Thereafter, this thesis compares the COMSOL simulation results (xii) of two wires and four wire bioimpedance measurement confirming that the two wire bioimpedance can be (up to two) orders of magnitude higher than four wire method due to contribution from highly insulating stratum corneum layer. To quantify the contribution of each layer to the total bioimpedance, selectivity analysis is performed. The normalized derivatives of the 4-wire impedance Z4W with respect to the electrical parameters of the different layers (i.e., the dielectric constants εk and the conductivities σk of all the tissues) are computed. The selectivity and normalized derivatives at various dimensions are crucial to design epidermal electrodes that could measure the electrical properties and bioimpedance of a single layer of interest. For instance, for the detection of cancer in viable skin, the system should ideally be able to measure, at a given frequency, the conductivity and/or the dielectric constant of viable skin. Electrical properties of skin layers are a strong function of frequency and bioimpedance is generally characterized over a frequency range from tens of Hz to mega Hz. Due to the frequencies involved, parasitic capacitances associated with the epidermal electrodes provide leakage current path and therefore act as a source of errors. This thesis models the parasitic capacitances associated with epidermal electrodes and computes the relative errors in skin’s bioimpedance measurement due to the patristics. The findings of this thesis can advance the development of bioimpedance systems for the characterization of deep bioimpedance for monitoring health and disease diagnosis of skin.