Protein is one of the fundamental macronutrients for the development and maintenance of the human body. In the last few years, the concern with the population increase and reduced resources, has generated a change in the consumption of animal protein for vegetable protein and emerging sources (e.g. insect). The protein isolated from different sources can be used for the development of new food formulations, as a wall material to encapsulate compounds of interest or as a biopolymer for packaging applications. Obtaining protein products involves several parameters: (1) quality of the protein isolate, (2) control of the production process of the materials, including the efficiency of the process, and (3) quality of the final product. In this approach, it is necessary to develop analysis methodologies that are fast, reliable, free of reagents and easy to execute. Hyperspectral images (HSI) fulfill this criterion, and their application in protein products is, until today, little explored. However, the large amount of information provided by HSI must be analyzed using Artificial Intelligence (AI), such as machine learning methods. In this project, we propose to use HSI in combination with AI to develop new analytical methodologies for the quality control of protein products, which includes protein isolates, oil-filled microcapsules and hydrogels and curcumin-loaded aerogels.
News published in Agência FAPESP Newsletter about the scholarship: