Taguchi Design of Experiment for Optimizing Saline Solution Composition to Increase Electrical Coupling in a Multi-Electrode External EEG Sensor

Abstract The effectiveness of noninvasive brain sensing lies in the development of EEG (electroencephalogram) sensors capable of delivering high signal-to-noise ratios for real-time brain-machine interfacing and neuromodulation. A critical factor for achieving this goal is the optimization of electrical coupling between the skin and external electrodes, especially when detecting weak brain signals in the millivolt to microvolt range. In this study, we employed a Taguchi Design of Experiment (DoE) methodology to optimize the composition of an electrolyte solution used to enhance signal quality in a 14-electrode wearable EEG sensor system. We investigated four electrolyte components Sodium Chloride (NaCl), Potassium Chloride (KCl), Sodium Bicarbonate (NaHCO 3 ), and Tween-20 surfactant each varied across three concentration levels using an L9 orthogonal array. For each experimental condition, ten measurements were collected to ensure statistical reliability. Two statistical analytical approaches were applied: average signal strength analysis, which exhibited considerable variability and limited diagnostic value, and signal-to-noise ratio (SNR) analysis, which used logarithmic scaling to provide more meaningful insights by minimizing noise effects. Results underscore that while the three salts made comparable contributions, the Tween-20 surfactant was the dominant factor affecting EEG signal strength. Additionally, all four factors exhibited non-linear behavior across concentration levels. This work presents a systematic optimization approach for improving electrolyte formulations and offers practical guidance for enhancing electrical coupling in EEG systems. The results support the futuristic development of high-SNR nanoscale EEG technologies where skin-sensor coupling will be highly sensitive towards saline solution as a conductive medium.