Machine Learning Enabled Reusable Adhesion, Entangled Network-Based Hydrogel for Long-Term, High-Fidelity EEG Recording and Attention Assessment

Abstract

Due to their high mechanical compliance and excellent biocompatibility, conductive hydrogels exhibit significant potential for applications in flexible electronics. However, as the demand for high sensitivity, superior mechanical properties, and strong adhesion performance continues to grow, many conventional fabrication methods remain complex and costly. Herein, we propose a simple and efficient strategy to construct an entangled network hydrogel through a liquid–metal-induced cross-linking reaction, hydrogel demonstrates outstanding properties, including exceptional stretchability (1643%), high tensile strength (366.54 kPa), toughness (350.2 kJ m−3), and relatively low mechanical hysteresis. The hydrogel exhibits long-term stable reusable adhesion (104 kPa), enabling conformal and stable adhesion to human skin. This capability allows it to effectively capture high-quality epidermal electrophysiological signals with high signal-to-noise ratio (25.2 dB) and low impedance (310 ohms). Furthermore, by integrating advanced machine learning algorithms, achieving an attention classification accuracy of 91.38%, which will significantly impact fields like education, healthcare, and artificial intelligence.

1. Introduction

Hydrogel epidermal electrodes play a crucial role in the transmission of electrophysiological signals [1–3]. In systems designed for the collection and processing of physiological signals, hydrogels serve as electrode materials, offering exceptional biocompatibility and effectively minimizing the risk of allergic reactions or rejection when in contact with human skin [4–6]. These epidermal electrodes, composed of conductive hydrogels, exhibit properties akin to natural skin, such as softness, moisture retention, and deformability [7, 8]. Consequently, they have garnered significant interest in the realm of flexible epidermal electronics, with applications ranging from artificial skin to physiological monitoring [9–11]. Hydrogel electrodes are capable of accurately capturing human motion and physiological signals, such as electrocardiograms (ECG) [12], electromyograms (EMG) [13, 14], and electroencephalograms (EEG) [15]. They are extensively utilized in diagnosing cardiovascular diseases, neurological therapies, health monitoring, and human–machine interaction fields [16–18]. While traditional wet electrodes, such as Ag/AgCl gel electrodes, can reduce interface impedance to some degree, but their prolonged use often lead to signal loss and complications, including localized skin sensitization, irritation, and rashes [19, 20]. Therefore, to facilitate the long-term and accurate collection of electrophysiological signals, it is imperative to develop flexible electrodes that address these limitations.

Gelatin is a biomolecule derived from the thermal denaturation of collagen, characterized by excellent biocompatibility, temperature sensitivity (26–30 °C), water solubility, adhesion, and cost-effectiveness [21, 22]. The adhesive properties of gelatin stem from the abundance of reactive groups present in its amino acid side chains, including amino, hydroxyl, carboxyl, and sulfhydryl groups [23, 24]. Leveraging these temperature-sensitive properties, Wang [21] and Li et al. [25] reported the development of skin-coatable biogel that facilitates in situ gelation to the skin interface during EEG acquisition and can be easily removed post-use. However, the mechanical properties and stability of this biogel are not optimal. Like other single-network hydrogels [26, 27], gelatin hydrogels typically display lower strength and poor stability, which significantly constrains their application in dynamic signal monitoring.

In recent years, novel polymer network strategies such as interpenetrating networks [28], nanocomposites [29, 30], and dual-network [31] structures have been reported. These typically consist of interpenetrating brittle and d