使用ATR-FTIR光谱和可解释的机器学习来预测和分类小麦粉中的脱氧尼瓦伦醇
Jiajun Liu1, Kebing Yao2, Chen Chen2
1Zhenjiang Academy of Agricultural Sciences/ Zhenjiang Agricultural Science Research Institute of Jiangsu Hilly Regions, Jiangsu Academy of Agricultural Sciences, Jurong, China; International Maize and Wheat Improvement Center (CIMMYT), Texcoco, Mexico.
富里埃变换红外光谱法 (FTIR) 与机器学习相结合,可以准确预测小麦中的脱氧尼瓦醇 (DON). 这种具有成本效益的方法可以通过可靠的DON水平监测来提高食品安全.
科学领域:
- 农业化学 农业化学
- 分析化学 分析化学
- 食品科学 食品科学 食品科学
背景情况:
- 脱氧尼瓦伦醇 (DON) 是小麦中重要的真菌毒素污染物,对食品安全构成风险.
- 目前用于DON量化的方法缺乏可靠性和成本效益.
- 菌头炎是小麦核中DON污染的主要原因.
研究的目的:
- 利用ATR-FTIR光谱学开发和验证一种具有成本效益的方法,用于预测小麦中的DON度.
- 为了比较四个机器学习模型对DON量化和分类的性能.
- 使用SHAP分析提高预测模型的可解释性.
主要方法:
- 在采集光谱数据时,采用了具有减弱总反射率 (ATR-FTIR) 的里埃变换红外光谱法.
- 四个机器学习模型 (XGBoost,RF,TabPFN,CatBoost) 在通过RFE选择的光谱特征上受过训练.
- 为了模型的可解释性,使用了夏普利添加式扩展 (SHAP).
主要成果:
- TabPFN模型实现了最高的定量预测准确性 (R2 = 0.86).
- SHAP分析确定了影响DON预测的关键波数:996,1083,1135和1574厘米-1.1.
- 对于DON级别的二进制分类,CatBoost模型表现出卓越的性能 (回忆=0.91,精度=0.88,F2=0.91).
结论:
- 对FTIR光谱的机器学习分析为在小麦中检测DON提供了有效的方法.
- 拟议的ATR-FTIR和ML方法提供了一个可行的解决方案,用于监测小麦面粉中的DON水平.
- 这种方法在实验室和工业环境中都有潜在的应用,以确保食品安全.
更多相关视频
04:09Predicting Treatment Response to Image-Guided Therapies Using Machine Learning: An Example for Trans-Arterial Treatment of Hepatocellular Carcinoma
Published on: October 10, 2018
07:13Comparison of Predictive Performance of Three Lymph Node Staging Systems in Colorectal Signet Ring Cell Carcinoma Based on Machine Learning Model
Published on: April 18, 2025
相关概念视频
Attenuated Total Reflectance (ATR) Infrared Spectroscopy: Overview
The ATR process begins by directing a beam...
Classifying Matter by Composition
According to its composition, the matter can be classified into two broad categories — pure substances and mixtures.
A pure substance is a form of matter that has a constant composition throughout with uniform properties. For example, any sample of sucrose has the same composition and same physical properties, such as melting point, color, and sweetness, regardless of the source from which it is isolated.
A mixture is composed of two or...
Classifying Matter by State
Predicting Molecular Geometry
Machines
A free-body diagram of the...
How Data are Classified: Numerical Data
Quantitative data may be either discrete or continuous. All quantitative data that take on only specific numerical...
