Related Experiment Video
Updated: Aug 5, 2026
09:23
Using Extraordinary Optical Transmission to Quantify Cardiac Biomarkers in Human Serum
Published on: December 13, 2017
Machine learning-assisted colorimetric sensor array on a superwettable microchip for on-site authentication of
Huan Jiang1, Bingxin Ma1, Xuyang Wang1
1Beijing Key Laboratory for Bioengineering and Sensing Technology, School of Chemistry and Biological Engineering, University of Science and Technology Beijing, Beijing, 100083, P.R. China.
Talanta
|August 1, 2026
Summary
This study introduces a smart sensor chip using machine learning for fast and accurate ginseng authentication. It effectively detects adulteration, ensuring the quality of ginseng products.
Area of Science:
- Analytical Chemistry
- Biotechnology
- Materials Science
Background:
- Panax ginseng is valued for bioactivities, but faces significant adulteration and counterfeiting challenges.
- Ensuring the authenticity and quality of ginseng products is crucial for consumer safety and efficacy.
Purpose of the Study:
- To develop a rapid, low-cost, and instrument-free method for ginseng authentication and quality control.
- To create a machine learning-assisted superwettable colorimetric sensor array microchip for detecting ginseng adulteration.
Main Methods:
- A colorimetric sensor array microchip was designed using 3-nitrophenylboronic acid to target diol groups in ginseng monosaccharides.
- The microchip employed pH response and indicator displacement assay (IDA) for colorimetric fingerprint generation.
- Machine learning algorithms, including support vector regression (SVR), were utilized for data analysis and prediction.
Main Results:
- The sensor array demonstrated excellent determination capabilities for classifying and semi-quantifying monosaccharides in ginseng hydrolysates.
- High accuracy was achieved in ginseng species identification (96.49%) and authenticity classification (99.02%) using machine learning.
- Support vector regression (SVR) accurately quantified adulteration ratios with a high predictive performance (R² = 0.9869).
Conclusions:
- The proposed machine learning-assisted sensor array provides a reliable and efficient strategy for on-site quality control of ginseng.
- This method offers a rapid, cost-effective, and instrument-free solution to combat ginseng adulteration and counterfeiting.
- The technology has been successfully validated for identifying commercial ginseng samples, ensuring product authenticity.
