Related Experiment Video
Updated: May 10, 2026
06:04
Source and Route of Pyrrolizidine Alkaloid Contamination in Tea Samples
Published on: September 28, 2022
1.8K
Deep Learning-Assisted Multiplexed Electrochemical Fingerprinting for Chinese Tea Identification
Yuyu Tan1, Mengli Luo1, Chao Xu1
1Hunan Province Key Laboratory for Ultra-Fast Micro/Nano Technology and Advanced Laser Manufacture, School of Electrical Engineering, University of South China, Hengyang 421001, China.
Analytical Chemistry
|April 10, 2025
Summary
A new laser-engraved sensor array uses electrochemical fingerprinting and a convolutional neural network (CNN) to accurately identify tea polyphenols and differentiate Chinese tea varieties. This method offers a reliable approach for agricultural product authentication.
Area of Science:
- Analytical Chemistry
- Sensor Technology
- Machine Learning
Background:
- Differentiating natural products with similar chemical structures in complex matrices is analytically challenging.
- Conventional methods often lack the sensitivity and specificity required for complex sample analysis.
- Accurate authentication of agricultural products like tea is crucial for quality control and consumer trust.
Purpose of the Study:
- To develop a novel sensing strategy for the rapid and precise detection of tea polyphenols.
- To differentiate various types of Chinese teas using multiplexed electrochemical fingerprinting.
- To enhance the accuracy of tea identification through advanced machine learning algorithms.
Main Methods:
- Development of a laser-engraved sensor array with three distinct working electrode configurations (bare, nanoenzyme, bioenzyme).
- Application of multiplex electrochemical fingerprinting to generate unique signals from complex samples.
- Utilization of a self-designed one-dimensional convolutional neural network (1D-CNN) for feature extraction and pattern recognition.
Main Results:
- Successful detection of three key tea polyphenols.
- Accurate differentiation of six Chinese tea series (98.84%) and 24 distinct tea varieties (97.68%).
- Demonstrated superior accuracy of the deep learning-assisted electrochemical fingerprinting compared to other machine learning methods.
Conclusions:
- The developed sensor array and 1D-CNN algorithm provide a rapid, reliable, and highly accurate method for tea polyphenol detection and tea authentication.
- This approach significantly advances the capabilities for identifying and authenticating complex agricultural products.
- The study highlights the potential of integrating advanced sensing technologies with deep learning for food analysis and quality assurance.
