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

Area of Science:

• Electrochemistry
• Biosensing
• Artificial Intelligence

Background:

• Accurate small molecule detection is vital across scientific disciplines.
• Traditional electrochemical methods often lack sufficient accuracy.
• Multi-mode sensors improve accuracy but struggle with signal deviation issues.

Purpose of the Study:

• To develop an advanced multi-mode sensor for ascorbic acid (AA) detection.
• To address sensor failure caused by deviations in multi-modal signal analysis.
• To enhance the prediction accuracy, detection range, and anti-interference capabilities of sensors.

Main Methods:

• Development of a multi-mode sensor utilizing Prussian Blue (PB) for AA detection.
• Innovative integration of back-propagation artificial neural networks (BP ANNs).
• Comprehensive processing of three distinct signal datasets using BP ANNs.

Main Results:

• Successfully mitigated sensor failure associated with large signal deviations.
• Significantly improved prediction accuracy for ascorbic acid detection.
• Enhanced the overall detection range and anti-interference performance of the sensor.

Conclusions:

• The proposed BP ANN integration offers an effective solution for multi-modal sensor data analysis.
• This approach overcomes critical limitations in existing multi-mode sensor designs.
• The findings demonstrate broad applicability in bioanalysis, clinical diagnosis, and related fields.