Back Propagation Artificial Neural Network Enhanced Accuracy of Multi-Mode Sensors

The detection of small molecules is critical in many fields, but traditional electrochemical detection methods often exhibit limited accuracy. The construction of multi-mode sensors is a common strategy to improve detection accuracy. However, most existing multi-mode sensors rely on the separate analysis of each mode signal, which can easily lead to sensor failure when the deviation between different mode results is too large. In this study, we propose a multi-mode sensor based on Prussian Blue (PB) for ascorbic acid (AA) detection. We innovatively integrate back-propagation artificial neural networks (BP ANNs) to comprehensively process the three collected signal data sets, which successfully solves the problem of sensor failure caused by the large deviation of signal detection results, and greatly improves the prediction accuracy, detection range, and anti-interference of the sensor. Our findings provide an effective solution for optimizing the data analysis of multi-modal sensors, and show broad application prospects in bioanalysis, clinical diagnosis, and related fields.