Enterprise Europe Network

Assessing the degree of authenticity of natural flavourings

Country of origin:
Country: 
SLOVENIA
Opportunity:
External Id: 
TOSI20200703001
Published
03/07/2020
Last update
06/07/2020
Expiration date
07/07/2021

Keywords

Partner keyword: 
Artificial Intelligence (AI)
Knowledge Management, Process Management
Analytical Chemistry
Detection and Analysis methods
Traceability of food
Expert systems
Health food
Other Industrial Products (not elsewhere classified)
Agriculture, Forestry, Fishing, Animal Husbandry & Related Products
Other research and experimental development on natural sciences and engineering
EXPRESS YOUR INTEREST

Summary

Summary: 
The Slovenian public research organization has developed a new system for authenticity assessment of natural flavourings. This system is based on the analysis of specific compounds and their stable carbon isotope composition. The measurement data are analysed by machine-learning algorithms to characterise the authenticity of natural vs. artificial flavourings. Partners are sought among food and aroma industry for technical and research cooperation agreements.

Description

Description: 

The expanding industrialization of food and aroma products have generated the need to authenticate the production method of marketed products. With increasing pressure to satisfy consumer needs and the high price of natural flavourings compared to synthetic ones, makes naturally flavoured products especially vulnerable to economically motivated adulteration. The relatively new science of food forensics is employing a range of developing isotopic techniques that have allowed detecting adulterated and counterfeit food.

The Slovenian research group has developed a new system for authenticity assessment of natural flavourings. Typically, the artificial aroma is produced by mixing several different compounds originating mostly from oil. In essence, they are chemically very similar to natural compounds and usually being added to natural products makes them even more difficult to identify. The complexity of the problem has led to the development of a system based on machine-learning.
The basis of the system is a database describing different aromas, both authentic and artificial. The data are based on chemical analysis of compounds in a sample obtained by several methods: gas chromatography/mass selective detector (GC-MSD), gas chromatography—combustion—isotope ratio spectrometry (GC-C-IRMS) or elemental analyser/isotope ratio mass spectrometry (EA/IRMS).
The data obtained along with the expert knowledge are collected in a database and analysed. The analysis is based on the degree of confidence in the partial characterisation of the authenticity for any given flavour. It is a multi-step procedure, where each next step is driven by the previously obtained estimates.

The researchers come from a Slovenian public research organization. Their multi-disciplinary research involves a combination of physical, chemical and biology processes influencing the environment. On the other hand, there is a group of computer and artificial intelligence specialists who contributed to the flavours-characterization system based on data-mining approaches.

The partners are sought in food and aroma industry for technical cooperation agreements for sample testing and verification.

Additionally specialized laboratories for testing the food products are sought under research cooperation agreements for expanding the database and expert knowledge in classifying aroma compounds.

Advantages & innovations

Cooperation plus value: 
Aroma classifications are performed by machine learning rather than manually, by an expert.

Stage of development

Cooperation stage dev stage: 
Available for demonstration

Partner sought

Cooperation area: 
Partners are sought among food and aroma industry for technical cooperation agreements for random sample testing applications Partners are sought among specialized laboratories for testing food products under research cooperation agreements for expanding the database and expert knowledge in classifying aroma compounds.

Type and size

Cooperation task: 
SME 11-50,>500 MNE,251-500,SME 51-250,>500