Revolutionising Flavour: Nature-Inspired and AI-Driven

One of the main reasons for the low growth rate of plant-based meat and dairy alternatives is the difference in sensory properties compared to their meat and dairy counterparts. The ALL-INFORMED project aims to revolutionize the process development of meat and dairy alternatives by using smart technologies and digital innovation. We aim to expand the application of traditional processes by predicting the right process for obtaining plant-based products with an improved sensory experience.

Use of Artificial Intelligence

The amount of data provided by analytical tools of Near Infrared (NIR) and High-Resolution Mass Spectrometry (HRMS) exceeds the human ability for interpretation, hence the use of Artificial Intelligence (AI) is an essential part of this project. We will use AI to combine and interpret the vast information obtained from combinatorial fermentations (1000s of micro-organisms and 10s of substrates), reduce off-notes and form specific aroma molecules related to dairy and meaty products.

Predicting plant-based flavour aromas

The ALL-INFORMED project aims to develop an AI tool that utilizes machine learning algorithms to predict natural, clean label, plant-based flavour aromas and their precursors from the fermentation of plant proteins and food processing side streams. The tool will help guide the fermentation process development of plant-based products such as meat analogues and dairy alternatives by combining information from non-invasive sensor technologies and high-throughput HRMS.

A novel standard on process design and understanding

The data obtained will be organized and structured using semantic technologies adhering to the principles of Findability, Accessibility, Interoperability and Reusability (FAIR). The AI tool will combine access and exploration of different datasets and the implementation of a predictive machine learning algorithm for aroma flavour prediction from fermentation. By using many different substrates and cultures, we will strengthen the predictive power of the AI tool and its algorithms while keeping specific sensitive company data uncompromised.

The end result will be a digital platform that can be used to predict the right combination of strains and substrates for a flavour (aroma) profile that mimics the authentic sensory profile of meat and dairy products. This project will establish a novel standard on process design and understanding and, at the same time, showcase the power of AI in food processing.