Insect-based powder detection in plant-based powder foods by spectroscopy and hyperspectral imaging combined with chemometrics

• Grant number: 23/11773-2
• Support Opportunities: Scholarships abroad - Research Internship - Doctorate
• Start date: February 29, 2024
• End date: October 28, 2024
• Field of knowledge: Agronomical Sciences - Food Science and Technology - Food Engineering
• Principal Investigator: Douglas Fernandes Barbin
• Grantee: Luis Jam Pier Cruz Tirado
• Supervisor: Christopher Trevor Elliott
• Host Institution: Faculdade de Engenharia de Alimentos (FEA). Universidade Estadual de Campinas (UNICAMP). Campinas , SP, Brazil
• Institution abroad: Queen's University Belfast, Northern Ireland
• Associated to the scholarship: 20/09198-1 - Hyperspectral imaging and artificial intelligence for quality control of protein-based products: isolates, microcapsules and gels, BP.DR

Abstract

The increase in the world's population compels us to search for new alternative protein sources, such as insect protein. However, despite the nutritional and environmental benefits associated with insect production and consumption, Western cultures exhibit a certain level of rejection mainly due to knowledge, emotional factors, neophobia, taste, and texture. Nonetheless, currently, commercially available insects for human consumption, such as crickets (Acheta domesticus) and lesser mealworms (Alphitobius diaperinus), are available, with the anticipation of the market diversifying further by including new insects as novel food in the future. Therefore, it is essential for regulatory and research organizations to promote the development of analytical techniques that can detect the purity of the protein sources we consume: insect-based or plant-based powder. In this regard, this project aims to develop analytical models using information obtained from a portable NIR spectrometers and hyperspectral imaging (NIR-HSI). Predictive models will be developed using traditional chemometric models such as Partial Least Square Discriminant Analysis (PLS-DA) and regression (PLSR), Support Vector Machine Discriminant Analysis (SVM-DA) and regression (PLSR), and Data Driven Soft Independent Class Analogy (DD-SIMCA). It is expected that the results will demonstrate the excellent performance of the portable NIR spectrometer and NIR-HSI in authenticating pure insect-based and plant-based powder and blends, as well as quantifying the level of mixture of both powders. Both techniques would offer two viable options for quality control of these protein sources and can be implemented as screening methods, complementing each other effectively.