Fast online estimation of quail eggs freshness using portable NIR spectrometer and machine learning

Author(s)

Brasil, Yasmin Lima ^[1] ; Cruz-Tirado, J. P. ^[1] ; Barbin, Douglas Fernandes ^[1]

Affiliation

^[1] Univ Estadual Campinas, Dept Food Engn, Rua Monteiro Lobato, 80, Cidade Univ, BR-13083862 Campinas, SP - Brazil

• Document type: Journal article
• Source: FOOD CONTROL; v. 131, JAN 2022.
• Web of Science Citations: 3

Abstract

Quail eggs are one of the main natural sources of essential nutrients, presenting high amounts of protein, antioxidants, calcium, iron and phosphorus. However, its quality assessment demands laborious methods and chemicals, and there is currently no standard method do quantify its freshness. This work aimed to investigate the performance of a portable NIR spectrometer, in combination with machine learning, to estimate the freshness of quail eggs. Since there is no standard index to classify quail eggs, we compared Haugh Unit (HU), Yolk Index (YI) and the Egg Quality Index (EQI) as reference methods. Partial Least Squares Regression (PLSR) and Support Vector Machine Regression (SVMR) were used to build prediction models, and Partial Least Squares-Discriminant Analysis (PLSDA) and Support Vector Machine Classification (SVMC) for the development of classification models. For the first time, we demonstrated that EQI, which is a parameter that measures egg freshness according to the quality of the yolk and the albumen, is the best way to express the freshness of quail eggs. The best prediction models were obtained for YI and EQI, using SVMR, with RPD = 2.0-2.5 and RER >10, indicating good predictive capacity. PLSDA and SVMC models showed similar performance, correctly classifying more than 80% of the samples. The results obtained demonstrate the potential of portable NIR spectrometer for monitoring quail eggs freshness during storage. (AU)

• FAPESP's process: 20/09198-1 - Hyperspectral imaging and artificial intelligence for quality control of protein-based products: isolates, microcapsules and gels
• Grantee: Luis Jam Pier Cruz Tirado
• Support Opportunities: Scholarships in Brazil - Doctorate
• FAPESP's process: 15/24351-2 - Applications of image analyses and NIR spectroscopy for quality assessment and authentication of food products
• Grantee: Douglas Fernandes Barbin
• Support Opportunities: Research Grants - Young Investigators Grants
• FAPESP's process: 18/02500-4 - Food analyses using NIR spectral imaging
• Grantee: Luis Jam Pier Cruz Tirado
• Support Opportunities: Scholarships in Brazil - Master
• FAPESP's process: 19/11896-1 - Multivariate statistical analyses applied to NIR spectroscopy and digital image analyses for food products
• Grantee: Yasmin Lima Brasil
• Support Opportunities: Scholarships in Brazil - Master