AI/ML-Enabled 2-D - RuS2 Nanomaterial-Based Multifunctional, Low Cost, Wearable Sensor Platform for Non-Invasive Point of Care Diagnostics

Abstract

We report the first of its kind artificial intelligence/machine learning (AI/ML) enabled nanomaterial based multifunctional sensing platform for simultaneous and continuous monitoring of certain vital body parameters viz. the hydration levels of the skin, glucose concentration and pH levels in biofluid sweat with high accuracy and speed. RuS 2 nanoparticles were synthesized using a facile hydrothermal method and detailed characterization revealed cubic crystal structure of laurite-RuS 2 with mesoporous morphology that provided enhanced electrocatalytic sites for sensing. The biochemical sensor was fabricated using layer-by-layer spincoating technique of RuS 2 on an interdigitated PDMS substrate. Further, to facilitate human-machine interface that can analyze data from large sample sizes (~ 50 sensors), the sensor was interfaced with the open-source microcontroller board (QueSSence) wherein Artificial Intelligence (AI) based K-Nearest Neighbors (KNN) algorithm enabled precise and faster data acquisition from complex mathematical conjunctures. The response of the sensor towards hydration levels for various skin conditions, glucose concentration, and pH of sweat were examined for both artificial skin and human skin. The RuS 2 based sweat-glucose sensor exhibited a sensitivity of 87.9 ± 0.6 μM -1 cm -2 in a physiologically relevant range of 10 nM - 0.1 mM and limit of detection of 4.87 nM. The pH sensor exhibited a sensitivity of 71.2 ± 0.5 pH -1 cm -2 across the pH range of 4 -8.5. The multifunctional sensor displayed high stability and reusability at room temperature. The optimized response was integrated with a smartphone via a customized application enabling user-friendly, real-time monitoring of the health conditions.