Detection of the 5-hydroxymethylfurfural content in roasted coffee using machine learning based on near-infrared spectroscopy

Chuanqi Xie; Changyan Wang; Mengyao Zhao; Weidong Zhou

doi:10.1016/j.foodchem.2023.136199

Detection of the 5-hydroxymethylfurfural content in roasted coffee using machine learning based on near-infrared spectroscopy

Food Chem. 2023 Oct 1:422:136199. doi: 10.1016/j.foodchem.2023.136199. Epub 2023 Apr 20.

Authors

Chuanqi Xie¹, Changyan Wang², Mengyao Zhao³, Weidong Zhou⁴

Affiliations

¹ State Key Laboratory for Managing Biotic and Chemical Threats to the Quality and Safety of Agro-Products, The Institute of Animal Husbandry and Veterinary Science, Zhejiang Academy of Agricultural Sciences, Hangzhou 310021, China.
² State Key Laboratory of Bioreactor Engineering, School of Biotechnology, East China University of Science and Technology, Shanghai 200237, China.
³ State Key Laboratory of Bioreactor Engineering, School of Biotechnology, East China University of Science and Technology, Shanghai 200237, China. Electronic address: myzhao@ecust.edu.cn.
⁴ State Key Laboratory for Managing Biotic and Chemical Threats to the Quality and Safety of Agro-Products, The Institute of Animal Husbandry and Veterinary Science, Zhejiang Academy of Agricultural Sciences, Hangzhou 310021, China. Electronic address: zhouwd@zaas.ac.cn.

PMID: 37121208
DOI: 10.1016/j.foodchem.2023.136199

Abstract

Since 5-hydroxymethylfurfural (5-HMF) is carcinogenic to humans, its detection in foods is essential. This study performed near-infrared (NIR) spectroscopy (11998-4000 cm^-1) to determine the 5-HMF content in roasted coffee. The random forest (RF) was used to extract important wavenumbers, after which three machine learning models (ordinary least square (OLS), support vector machine (SVM), and RF) were established for the prediction. RF obtained the best prediction results (R_c² = 0.98 and R_p² = 0.92) compared with OLS and SVM and effectively extracted the important wavenumbers (11667 cm^-1, 11666 cm^-1, 10905 cm^-1, 7096 cm^-1, 7095 cm^-1, 7094 cm^-1, 7093 cm^-1, 7092 cm^-1, 5054 cm^-1, 5026 cm^-1, 5025 cm^-1, and 5024 cm^-1). The results demonstrated that machine learning models based on NIR spectroscopy could provide a non-destructive approach for determining 5-HMF content in roasted coffee.

Keywords: 5-hydroxymethylfurfural; Content prediction; Machine learning; NIR spectroscopy; Roasted coffee.

MeSH terms

Coffee* / chemistry
Humans
Least-Squares Analysis
Seeds / chemistry
Spectroscopy, Near-Infrared* / methods
Support Vector Machine

Substances

Coffee
5-hydroxymethylfurfural