Development and validation of predictive models based on a portable spectrometer and NIR hyperspectral imaging and machine learning to predict the composition of black soldier fly larvae (Hermetia illucens)

• Grant number: 22/07725-0
• Support Opportunities: Scholarships in Brazil - Scientific Initiation
• Start date: December 01, 2022
• End date: November 30, 2024
• Field of knowledge: Physical Sciences and Mathematics - Chemistry - Analytical Chemistry
• Principal Investigator: Douglas Fernandes Barbin
• Grantee: Matheus Silva dos Santos Vieira
• Host Institution: Faculdade de Engenharia de Alimentos (FEA). Universidade Estadual de Campinas (UNICAMP). Campinas , SP, Brazil

Abstract

With the current population growth, the food industry is looking for sustainable sources of food. Edible insects are seen as a sustainable source of protein and fat, which can help fill the world's food deficit. In this group, the larvae of the black soldier fly (Hermetia illucens) has aroused great interest for being able to quickly (~14 days) transform organic waste into high quality protein, as well as fat, chitin and minerals. The insect production industry is growing in the world, including Brazil. Therefore, it is necessary to develop methodologies for analyzing the quality of larvae and larvae flour quickly and efficiently. Near-infrared spectroscopy (NIRS) allows the development of a non-destructive analysis methodology, fast and with a minimum of sample preparation. Portable NIR spectrometers have emerged as a way to perform in situ measurements due to their ergonomic design, ease of transport and relatively cheap cost. On the other hand, NIR hyperspectral images simultaneously provide spectral (chemical) and spatial (physical) information for the entire region of interest without the need for contact with the sample. For these reasons, they can be an alternative to traditional methods, allowing better quality control of black soldier fly larvae. Thus, this project aims at the development and validation of predictive models based on portable NIR spectrometers or hyperspectral images and machine learning to predict the chemical composition of black soldier fly larvae, more specifically fat content, protein content and fatty acid profile.

Scientific publications

(References retrieved automatically from Web of Science and SciELO through information on FAPESP grants and their corresponding numbers as mentioned in the publications by the authors)

CRUZ-TIRADO, J. P.; VIEIRA, MATHEUS SILVA DOS SANTOS; AMIGO, JOSE MANUEL; SICHE, RAUL; BARBIN, DOUGLAS FERNANDES. Prediction of protein and lipid content in black soldier fly (Hermetia illucens L.) larvae flour using portable NIR spectrometers and chemometrics. FOOD CONTROL, v. 153, p. 12-pg., 2023-11-01. (14/50951-4, 20/09198-1, 15/24351-2, 08/57808-1, 22/07725-0)