Abstract

Tear analysis plays a crucial role in the early screening and diagnosis of ophthalmic diseases. However, conventional methods are often limited by poor real-time performance, low portability, and insufficient capability for multi-parameter detection. Here, we present a bioinspired triboelectric electronic tongue integrated with a microfluidic chip for multimodal detection of tear pH and disease-related biomarkers. The system combines three triboelectric nanogenerator (TENG) modes, including droplet-based, dual-electrode sliding, and single-electrode sliding configurations. The droplet-based TENG converts gravitational potential energy into electrical energy, generating a maximum output voltage of 65 V. The sliding TENG further expands the sensing dimensions by characterizing droplet flow behavior and viscosity-related properties. Benefiting from the high sensitivity of the dual-electrode mode and the waveform differentiation capability of the single-electrode mode, the platform enables enhanced sample discrimination. After optimizing key parameters, including droplet height and chip inclination angle, the output stability errors for all three TENG modes were maintained within ±10%. Combined with a random forest algorithm, the multimodal sensing system achieved a classification accuracy exceeding 96.6% for artificial tears with different pH values. Moreover, distinct electrical response patterns were observed for ophthalmic disease-related biomarkers, including Lysozyme, Interleukin-6 (IL-6), and Chlamydia, demonstrating excellent type identification and concentration detection capability. This work provides a self-powered and miniaturized strategy for intelligent tear analysis and multiple-parameter sensing, offering significant potential for ophthalmic disease diagnosis.

IPC Classification

Keywords