Design and Implementation of Low-Cost Dehydration Sensor for Wearable Health Technology

Body hydration is considered one of the most important physiological parameters to measure, and one of the most challenging. Current methods to assess hydration are invasive and require costly clinical settings. Given the increased attention to wearable devices in healthcare, bio-impedance analysis offers a non-invasive and inexpensive tool to assess hydration, and it can be designed to be used in wearable health devices. In this thesis, the utilization of the bio-impedance analysis to be used in the design of a dehydration sensor is investigated. A model of the human skin is used to simulate and study the bio-impedance analysis method. Several simulation setups were implemented to study different factors including, the effect of applying different frequencies on the dielectric parameters of the skin, the effect of changing sensor spacing, and the effect of temperature on the resulting measurements. Two different interdigitated electrodes designs are presented and a comparison of the measurements is presented. The rectangular IDE is printed and tested on subjects to validate the bio-impedance method and study the interpretation of its results. The proposed design offers a classification criterion that can be used to assess dehydration without the need for a complex mathematical model. Further clinical testing and data are needed to refine and finalize the criteria.