Electrodermal Activity Sensor (EDA) with Adaptive Gain Control

António Francisco Rodrigues Banganho

Key information

Authors:

António Francisco Rodrigues Banganho (António Francisco Rodrigues Banganho)

Supervisors:

Hugo Humberto Plácido da Silva (Hugo Humberto Plácido da Silva); Marcelino Bicho dos Santos (Marcelino Bicho dos Santos)

Published in

11/26/2019

Abstract

Electrodermal activity (EDA) is one of the physiological signals that can be measured unintrusively and associated with the sympathetic nervous system, which regulates the fight or flight response in humans. Although not a substitute for other biomedical signals, it has been successfully used, for example, in the study of psychophysiological conditions such as autism, epilepsy or burnout syndrome. However, there are still some hardware limitations, such as sensitivity to artefacts, variability by effect of environmental variables and recording locations, among others. This paper aims to solve some of the problems with EDA measurements by developing a new circuit with adaptive gain control, allowing to be sensitive according to the intra-variability found in a population, while keeping the overall cost low. It also studies electrodes composed of unconventional materials such as MEDTEX P130, GMLC material, or the possibility of using PLA composite electrodes produced by desktop 3D printing, as well as alternative locations to record EDA signal such as the forehead. The proposed circuit showed comparable results to that obtained with a reference device, obtaining a correlation coefficient of 0.967, and showed improved performance in cases where the comparable sensor would saturate. The proposed circuit also includes a new frontend designed to separate the two components, tonic and phasic, that compose the EDA signal. Results from unconventional electrode materials show that the MEDTEX P130 has comparable performance to that of reference material, with a correlation coefficient of 0.789, while tests performed on the forehead were not conclusive.