Yi-Ming Cao, Yan Zhang, Qi Wang, Ran Zhao, Mingxi Hou, Shuang-Ting Yu, Kai-Kuo Wang, Ying-Jie Chen, Xiao-Qing Sun, Shijing Liu, Jiong-Tang Li
{"title":"Skin hyperspectral imaging and machine learning to accurately predict the muscular poly-unsaturated fatty acids contents in fish.","authors":"Yi-Ming Cao, Yan Zhang, Qi Wang, Ran Zhao, Mingxi Hou, Shuang-Ting Yu, Kai-Kuo Wang, Ying-Jie Chen, Xiao-Qing Sun, Shijing Liu, Jiong-Tang Li","doi":"10.1016/j.crfs.2024.100929","DOIUrl":null,"url":null,"abstract":"<p><p>The polyunsaturated fatty acids (PUFAs), particularly eicosapentaenoic acid (EPA) and docosahexaenoic acid (DHA), are critical determinants of the nutritional quality of fish. To rapidly and non-destructively determine the muscular PUFAs in living fish, an accuracy technique is urgently needed. In this study, we combined skin hyperspectral imaging (HSI) and machine learning (ML) methods to assess the muscular PUFAs contents of common carp. Hyperspectral images of the live fish skin were acquired in the 400-1000 nm spectral range. The spectral data were preprocessed using Savitzky-Golay (SG), multivariate scattering correction (MSC), and standard normal variable (SNV) methods, respectively. The competitive adaptive reweighted sampling (CARS) method was applied to extract the optimal wavelengths. With the skin spectra of fish, five ML methods, including the extreme learning machine (ELM), random forest (RF), radial basis function (RBF), back propagation (BP), and least squares support vector machine (LS-SVM) methods, were used to predict the PUFAs and EPA + DHA contents. With the spectral data processed with the SG, the RBF model achieved outstanding performance in predicting the EPA + DHA and PUFAs contents, yielding coefficients of determination (R<sup>2</sup> <sub>P</sub>) of 0.9914 and 0.9914, root mean square error (RMSE) of 0.3352 and 0.3346, and mean absolute error (MAE) of 0.2659 and 0.2660, respectively. Finally, the visualization distribution maps under the optimal model would facilitate the direct determination of the fillet PUFAs and EPA + DHA contents. The combination of skin HSI and the optimal ML method would be promising to rapidly select living fish having high muscular PUFAs contents.</p>","PeriodicalId":10939,"journal":{"name":"Current Research in Food Science","volume":"9 ","pages":"100929"},"PeriodicalIF":6.2000,"publicationDate":"2024-11-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC11612356/pdf/","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"Current Research in Food Science","FirstCategoryId":"97","ListUrlMain":"https://doi.org/10.1016/j.crfs.2024.100929","RegionNum":2,"RegionCategory":"农林科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"2024/1/1 0:00:00","PubModel":"eCollection","JCR":"Q1","JCRName":"FOOD SCIENCE & TECHNOLOGY","Score":null,"Total":0}
引用次数: 0
Abstract
The polyunsaturated fatty acids (PUFAs), particularly eicosapentaenoic acid (EPA) and docosahexaenoic acid (DHA), are critical determinants of the nutritional quality of fish. To rapidly and non-destructively determine the muscular PUFAs in living fish, an accuracy technique is urgently needed. In this study, we combined skin hyperspectral imaging (HSI) and machine learning (ML) methods to assess the muscular PUFAs contents of common carp. Hyperspectral images of the live fish skin were acquired in the 400-1000 nm spectral range. The spectral data were preprocessed using Savitzky-Golay (SG), multivariate scattering correction (MSC), and standard normal variable (SNV) methods, respectively. The competitive adaptive reweighted sampling (CARS) method was applied to extract the optimal wavelengths. With the skin spectra of fish, five ML methods, including the extreme learning machine (ELM), random forest (RF), radial basis function (RBF), back propagation (BP), and least squares support vector machine (LS-SVM) methods, were used to predict the PUFAs and EPA + DHA contents. With the spectral data processed with the SG, the RBF model achieved outstanding performance in predicting the EPA + DHA and PUFAs contents, yielding coefficients of determination (R2P) of 0.9914 and 0.9914, root mean square error (RMSE) of 0.3352 and 0.3346, and mean absolute error (MAE) of 0.2659 and 0.2660, respectively. Finally, the visualization distribution maps under the optimal model would facilitate the direct determination of the fillet PUFAs and EPA + DHA contents. The combination of skin HSI and the optimal ML method would be promising to rapidly select living fish having high muscular PUFAs contents.
期刊介绍:
Current Research in Food Science is an international peer-reviewed journal dedicated to advancing the breadth of knowledge in the field of food science. It serves as a platform for publishing original research articles and short communications that encompass a wide array of topics, including food chemistry, physics, microbiology, nutrition, nutraceuticals, process and package engineering, materials science, food sustainability, and food security. By covering these diverse areas, the journal aims to provide a comprehensive source of the latest scientific findings and technological advancements that are shaping the future of the food industry. The journal's scope is designed to address the multidisciplinary nature of food science, reflecting its commitment to promoting innovation and ensuring the safety and quality of the food supply.