Highly Stable Polymeric Electrooculography Electrodes for Contactless Human-Machine Interactions

Abstract

Capturing the electrooculography (EOG) signals is very attractive for assistive devices and user interfaces for virtual reality (VR) systems. However, the current EOG acquisition systems face challenges in ensuring user comfort, particularly in terms of electrode electrical and mechanical performance, long-term usability, thermal effects, and overall system portability. This study presents polymeric dry flexible electrodes, composed of a composite of poly(3,4-ethylenedioxythiophene):polystyrenesulfonate (PEDOT:PSS), poly(vinyl alcohol) (PVA), Gallic acid (GA), and D-sorbitol, forming a dynamic cross-linked network that ensures strong adhesion, stretchability, and electrical stability. These electrodes maintain their performance for up to 72 h, and can be restored through heat reactivation if performance degrades after prolonged storage. This electrode exhibits excellent biocompatibility, causing no skin irritation or thermal effects with continuous use. We have also developed a flexible circuit for real-time signal processing and wireless transmission, which operates in coordination with the EOG electrodes. The system employs a convolutional neural network (CNN) to achieve a 97.1% accuracy in classifying various eye movement patterns. The system enables contactless control of digital interfaces through simple eye movements, offering a solution for long-term, comfortable, and high-fidelity EOG-based human-machine interfaces, particularly for VR integration and assistive technologies for individuals with disabilities.