Reusable Ag‐Nanorod Biochip and Machine Learning‐Enhanced Electrochemical Platform for Accurate Glucose Detection in Human Biofluids

Electrochemical biosensors, owing to their distinctive advantages like short detection time, small scale, cost‐effectiveness, and high detection sensitivity and selectivity, can be applied towards qualitative and quantitative analysis of various target analytes. However, enzyme immobilisation over the electrode surface, commonly used to enhance signal generation, presents disadvantages like increased sensor intricacy, decreased stability, and increased time consumption for sensor development. To address these issues, enzyme immobilisation‐free methods directly detecting target analytes offer higher sensor specificity, sensitivity, and selectivity within a homogeneous matrix. This study proposes an immobilisation‐free sensor for glucose detection using reusable screen‐printed biochip combined with long silver nanorods (Ag‐LNR). The fabricated nanozyme‐based biosensor was evaluated, and a good selectivity was observed for glucose detection with a low detection limit of 34.9 mg/dL and a linear range from 50 to 200 mg/dL. Further, combining machine learning provides rapid glucose prediction with high precision and accuracy. The developed sensor was successfully applied to monitor glucose levels in human biofluids, offering a promising solution for continuously monitoring glucose and other target analytes. In addition, using the machine learning approach with the help of logarithmic regression analysis helped achieve highly precise and accurate diabetes prediction.