Related Experiment Video
Updated: Aug 5, 2026
07:10
Untargeted Liquid Chromatography-Mass Spectrometry-Based Metabolomics Analysis of Wheat Grain
Published on: March 13, 2020
Synergistic microfluidic processing and feature selection for improved Raman detection of mycotoxins in wheat
Jingwen Zhu1, Junyu Wang1, Xianjun Sun1
1School of Electrical and Information Engineering, Jiangsu University, Zhenjiang 212013, PR China.
Summary
This study presents a novel method for detecting mycotoxins in wheat using microfluidics and advanced feature selection. The approach significantly reduces interference, enabling accurate ppb-level detection below regulatory limits.
Area of Science:
- Analytical Chemistry
- Food Safety
- Spectroscopy
Background:
- Mycotoxin contamination in wheat poses significant food safety risks.
- Accurate detection at ppb-levels is challenging due to matrix interference and weak signals.
Purpose of the Study:
- To develop a synergistic strategy for sensitive and selective Raman detection of mycotoxins in wheat.
- To overcome limitations of strong matrix interference and weak signals in trace analysis.
Main Methods:
- Utilized a polydimethylsiloxane microfluidic chip for toxin separation and enrichment based on solubility.
- Implemented a Monte Carlo sampling - weighted bootstrap sampling - least absolute shrinkage and selection operator (MC-WBS-LASSO) algorithm for feature selection.
- Constructed a partial least squares regression (PLSR) model for quantitative analysis.
Main Results:
- Effectively eliminated matrix interference from starch and proteins.
- Achieved estimated limits of detection below Chinese national regulatory limits (GB 2761-2017) for four mycotoxins.
- Reduced feature dimensionality from thousands to under one hundred, focusing on core functional group peaks.
- Obtained test set coefficients of determination exceeding 0.92 for all mycotoxins.
Conclusions:
- The synergistic microfluidic and feature selection strategy offers a robust solution for ppb-level mycotoxin detection.
- The method demonstrates high accuracy, stability, and efficiency, balancing analytical performance.
- This approach shows potential for standardization in detecting trace contaminants in food matrices.
