Abstract
[Objective] To establish a rapid and accurate detection and analysis method for mycotoxins. [Methods] Based on near-infrared spectroscopy, qualitative models for the contents of zearalenone (ZEN) and deoxynivalenol (DON) in maize were established.144 naturally contaminated maize samples were used, and near-infrared spectra were collected from two sample preparations, namely maize powder and toxin extract. Five pretreatment methods were applied to process the original spectra, and two characteristic wavelength screening methods were selected to further extract effective information from the spectra. Pollution classification models for ZEN and DON were established using three machine learning algorithms, namely k-nearest neighbor classification algorithm, least squares support vector machine, and random forest, as well as the long short-term memory (LSTM) network algorithm of deep learning. [Results] The LSTM algorithm outperforms other algorithms on the spectral data of maize powder. For ZEN, the classification accuracy of the test set of the optimal qualitative model reaches as high as 97 %, while that of the optimal qualitative model for DON is 83 %. [Conclusion] The LSTM algorithm can effectively alleviate the overfitting problem and significantly improve the classification performance of the model.
Publication Date
5-15-2026
First Page
59
Last Page
67
DOI
10.13652/j.spjx.1003.5788.2024.80820
Recommended Citation
Man, GAO; Chengjing, QIAN; Ziyuan, DING; Yunjing, LUO; and Chen, ZHAI
(2026)
"Qualitative analysis methods of ZEN and DON in maize based on near-infrared spectroscopy, traditional machine learning, and LSTM,"
Food and Machinery: Vol. 42:
Iss.
4, Article 7.
DOI: 10.13652/j.spjx.1003.5788.2024.80820
Available at:
https://www.ifoodmm.cn/journal/vol42/iss4/7
References
[1] HAO W,GUAN S,LI A P,et al.Mycotoxin occurrence in feeds and raw materials in China:a five-year investigation [J].Toxins (Basel),2023,15(1):63.
[2] JING S Y,LIU C M,ZHENG J,et al.Toxicity of zearalenone and its nutritional intervention by natural products [J].Food & Function,2022,13(20):10 374-10 400.
[3] HOU S L,MA J J,CHENG Y Q,et al.The toxicity mechanisms of DON to humans and animals and potential biological treatment strategies [J].Critical Reviews in Food Science and Nutrition,2023,63(6):790-812.
[4] 苏青云,齐莎日娜,雷元培,等.2022 年中国饲料和原料中霉菌毒素污染调查报告 [J].饲料工业,2024,45(9):121-125.SU Q Y,QI S R N,LEI Y P,et al.A survey report on the mycotoxin contamination of Chinese raw materials and feed in 2022 [J].Feed Industry,2024,45(9):121-125.
[5] CAGLAYAN M O,ŞAHIN S,ÜSTÜNDAĞ Z.Detection strategies of zearalenone for food safety:a review [J].Critical Reviews in Analytical Chemistry,2022,52(2):294-313.
[6] SZÉKÁCS A.Mycotoxins as emerging contaminants.Introduction to the special issue "rapid detection of mycotoxin contamination" [J].Toxins (Basel),2021,13(7):475.
[7] LIJALEM Y G,GAB-ALLAH M A,CHOI K,et al.Development of isotope dilution-liquid chromatography/tandem mass spectrometry for the accurate determination of zearalenone and its metabolites in corn [J].Food Chem,2022,384:132483.
[8] TYSKA D,MALLMANN A,GRESSLER L T,et al.Near-infrared spectroscopy as a tool for rapid screening of deoxynivalenol in wheat flour and its applicability in the industry [J].Food Additives & Contaminants:Part A,2021,38(11):1 958-1 968.
[9] FREITAG S,SULYOK M,LOGAN N,et al.The potential and applicability of infrared spectroscopic methods for the rapid screening and routine analysis of mycotoxins in food crops [J].Comprehensive Reviews in Food Science and Food Safety,2022,21(6):5 199-5 224.
[10] KOS G,SIEGER M,MCMULLIN D,et al.A novel chemometric classification for FTIR spectra of mycotoxin-contaminated maize and peanuts at regulatory limits [J].Food Additives & Contaminants:Part A,2016,33(10):1 596-1 607.
[11] ELEM T C,PIACENTINI K C,ALVES L T,et al.NIR spectroscopy and chemometric tools to identify high content of deoxynivalenol in barley [J].Food Additives & Contaminants:Part A,2020,37(9):1 542-1 552.
[12] DE GIROLAMO A,CERVELLIERI S,CORTESE M,et al.Fourier transform near-infrared and mid-infrared spectroscopy as efficient tools for rapid screening of deoxynivalenol contamination in wheat bran [J].Journal of the Science of Food and Agriculture,2019,99(4):1 946-1 953.
[13] 邓志扬,廖强,邵淑娟,等.基于自然语言处理的山楂果实品种 近 红 外 无 损 鉴 别 方 法 [J].食 品 工 业 科 技,2023,44(22):249-256.DENG Z Y,LIAO Q,SHAO S J,et al.Nondestructive near-infrared identification of hawthorn fruit cultivars based on natural language processing [J].Science and Technology of Food Industry,2023,44(22):249-256.
[14] 卢诗扬,张雷蕾,潘家荣,等.拉曼光谱结合 LSTM 长短期记忆 网 络 的 樱 桃 产 地 鉴 别 研 究 [J].光 谱 学 与 光 谱 分 析,2021,41(4):1 177-1 181.LU S Y,ZHANG L L,PAN J R,et al.Study on the indetification of the geographical origin of cherries using Raman spectroscopy and LSTM [J].Spectroscopy and Spectral Analysis,2021,41(4):1 177-1 181.
[15] 何卓霖,李子琨,穆蕾,等.同位素内标—高效液相色谱—串联质谱法同时测定小麦和玉米中 19种真菌毒素 [J].中国粮油学报,2022,37(7):150-157.HE Z L,LI Z K,MU L,et al.Simultaneous determination of isotope internal standard-high performance liquid chromatography-tandem mass spectrometry 19 mycotoxins in wheat and corn [J].Journal of the Chinese Cereals and Oils Association,2022,37(7):150-157.
[16] MOKARI A,GUO S X,BOCKLITZ T.Exploring the steps of infrared (IR) spectral analysis:pre-processing,(classical) data modelling,and deep learning [J].Molecules,2023,28(19):|6 886.
[17] BI Y M,YUAN K L,XIAO W Q,et al.A local pre-processing method for near-infrared spectra,combined with spectral segmentation and standard normal variate transformation [J].Analytica Chimica Acta,2016,909:30-40.
[18] MARTENS H,STARK E.Extended multiplicative signal correction and spectral interference subtraction:new preprocessing methods for near infrared spectroscopy [J].Journal of Pharmaceutical and Biomedical Analysis,1991,9(8):625-635.
[19] 蔡海辉,周岭,史舟,等.利用 CARS-BPNN 模型的南疆枣园土壤有机质高光谱反演 [J].光谱学与光谱分析,2023,43(8):2 568-2 573.CAI H H,ZHOU L,SHI Z,et al.Hyperspectral inversion of soil organic matter in jujube orchard in southern Xinjiang using CARS-BPNN [J].Spectroscopy and Spectral Analysis,2023,43(8):2 568-2 573.
[20] 李跑,周骏,蒋立文,等.窗口竞争性自适应重加权采样策略的 近 红 外 特 征 变 量 选 择 方 法 [J].光 谱 学 与 光 谱 分 析,2019,39(5):1 428-1 432.LI P,ZHOU J,JIANG L W,et al.A variable selection approach of near infrared spectra based on window competitive adaptive reweighted sampling strategy [J].Spectroscopy and Spectral Analysis,2019,39(5):1 428-1 432.
[21] LI H D,LIANG Y Z,XU Q S,et al.Key wavelengths screening using competitive adaptive reweighted sampling method for multivariate calibration [J].Analytica Chimica Acta,2009,648(1):77-84.
[22] ZOU X B,ZHAO J W,POVEY M J W,et al.Variables selection methods in near-infrared spectroscopy [J].Analytica Chimica Acta,2010,667(1/2):14-32.[23] ZHANG S C,LI X L,ZONG M,et al.Efficient KNN classification with different numbers of nearest neighbors [J].IEEE Transactions on Neural Networks and Learning Systems,2018,29(5):1 774-1 785.
[24] MA Y F,LIANG X,SHENG G,et al.Noniterative sparse LS-SVM based on globally representative point selection [J].IEEE Transactions on Neural Networks and Learning Systems,2021,32(2):788-798.
[25] RIGATTI S J.Random forest [J].Journal of Insurance Medicine,2017,47(1):31-39.
[26] BECKER T,ROUSSEAU A J,GEUBBELMANS M,et al.Decision trees and random forests [J].American Journal of Orthodontics and Dentofacial Orthopedics,2023,164(6):894-897.
[27] MORO M K,DE CASTRO E V R,ROMÃO W,et al.Data fusion applied in near and mid infrared spectroscopy for crude oil classification [J].Fuel,2023,340:127580.
[28] FREITAG S,ANLANGER M,FOMINA P,et al.Attenuated total reflection mid-infrared spectroscopy to screen Austrian and French wheat from multiple years for deoxynivalenol [J].Spectrochimica Acta.Part A,Molecular and Biomolecular Spectroscopy,2025,340:126340.
[29] ALI H,MUTHUDOSS P,CHAUHAN C,et al.Machine learning-enabled NIR spectroscopy.Part 3:hyperparameter by design (hyd) based ANN-MLP optimization,model generalizability,and model transferability [J].AAPS Pharm Sci Tech,2023,24(8):254.
[30] KERNBACH J M,STAARTJES V E.Foundations of machine learning-based clinical prediction modeling:part II-generalization and overfitting [J].Acta Neurochirurgica,2022,134:15-21.
