基于荧光的传感器结合化学计量学和深度学习方法检测和定量牛乳中的椰奶掺假
A Fluorescence-Based Sensor Combined with Chemometric and Deep Learning Approaches for Detecting and Quantifying Coconut Milk Fraud in Bovine Milk.
摘要(中文译文)
牛乳中掺入椰奶对食品安全构成重大威胁，因为这两种液体在视觉上相似但营养上存在差异。本研究提出了一种集成分析框架，将激发-发射矩阵（EEM）荧光光谱与化学计量学和深度学习技术相结合，用于检测和定量牛乳中九个浓度水平（0-100%）的椰奶掺假。平行因子分析（PARAFAC）解析出两个主要荧光组分，即色氨酸（λ ex/em：290/350 nm）和核黄素（λ ex/em：450/525 nm），其得分随掺假程度增加而单调下降，证实了它们作为关键化学标志物的作用。在定量预测方面，利用三个激发波长下的发射光谱开发了PLSR和1D-CNN模型，最佳性能在450 nm处实现（PLSR：RP = 0.97，RMSEP = 5.00%；1D-CNN：RP = 0.94，RMSEP = 6.75%）。一种以完整EEM等高线图作为图像输入的轻量级2D-CNN优于所有定量模型，实现了RP = 0.99、RMSEP = 2.36%和RPD = 12.97，证明了保留完整二维荧光拓扑结构相较于离散波长选择的优势。这些结果证实，EEM结合2D-CNN为乳制品真伪鉴别提供了一种高精度且无损的工具。
Abstract
Bovine milk adulteration with coconut milk poses a significant threat to food safety, as both liquids are visually similar yet nutritionally distinct. This study presents an integrated analytical framework combining excitation-emission matrix (EEM) fluorescence spectroscopy with chemometric and deep learning techniques to detect and quantify coconut milk adulteration in bovine milk across nine concentration levels (0-100% /). Parallel factor analysis (PARAFAC) resolved two dominant fluorescent components, tryptophan (λ ex/em: 290/350 nm) and riboflavin (λ ex/em: 450/525 nm), whose scores decreased monotonically with increasing adulteration, confirming their role as key chemical biomarkers. For quantitative prediction, PLSR and 1D-CNN models were developed using emission spectra at three excitation wavelengths, with best performances achieved at 450 nm (PLSR: RP = 0.97, RMSEP = 5.00%; 1D-CNN: RP = 0.94, RMSEP = 6.75%). A lightweight 2D-CNN utilizing full EEM contour maps as image inputs outperformed all quantitative models, achieving RP = 0.99, RMSEP = 2.36%, and RPD = 12.97, demonstrating the advantage of preserving the full two-dimensional fluorescence topology over discrete wavelength selection. These results confirm that EEM combined with 2D-CNN provides a highly accurate and non-destructive tool for dairy authentication.
如何引用
AMA
Stella Maria Dyah Cahyarani, Hoonsoo Lee. A Fluorescence-Based Sensor Combined with Chemometric and Deep Learning Approaches for Detecting and Quantifying Coconut Milk Fraud in Bovine Milk.. Sensors (Basel, Switzerland). 2026; doi:10.3390/s26092872.
APA
Stella Maria Dyah Cahyarani, Hoonsoo Lee (2026). A Fluorescence-Based Sensor Combined with Chemometric and Deep Learning Approaches for Detecting and Quantifying Coconut Milk Fraud in Bovine Milk.. Sensors (Basel, Switzerland). https://doi.org/10.3390/s26092872
GB/T 7714
Stella Maria Dyah Cahyarani, Hoonsoo Lee. A Fluorescence-Based Sensor Combined with Chemometric and Deep Learning Approaches for Detecting and Quantifying Coconut Milk Fraud in Bovine Milk.[J]. Sensors (Basel, Switzerland), 2026 doi:10.3390/s26092872.
在 PubMed增强版 阅读全文
翻译全文 · AI 文献问答 · PDF 原文下载 · 引用导出