Accurate Subcutaneous Glucose Prediction Based on Reverse Iontophoresis with Skin Surface PH Calibration

Reverse iontophoresis (RI)-based approaches face challenges in achieving accurate epidermal-to-subcutaneous biomarkers prediction. Previously, we found that the skin surface pH undergoes significant change during RI, affecting the skin's zeta potential and, consequently, influencing the glucose extraction flux; and the changed pH also affect the performance of the glucose sensor. In this work, we firstly experimentally explore the impact of the change of skin surface pH on glucose extraction by RI and on the glucose sensor performance, thus then making accurate subcutaneous glucose prediction based on pH calibration. A flexible electronic device was developed, incorporating symmetric extraction electrodes for interstitial fluid (ISF) collection and sensors for glucose, Na<sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">+</sup> and pH measurement. The results showed that the device achieves a mean absolute percentage error (MAPE) of 2.9% in glucose prediction with pH calibration, while for 11.2% without calibration. This finding offers key insights into the RI mechanism of ISF extraction and related precise health monitoring.