Neural Network Chemical Compound Replacement for Plant-Based Foods

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Current methods for developing plant-based food alternatives struggle to match the taste and texture of animal-based foods, relying on inefficient and time-consuming manual laboratory processes, and often require expensive or hard-to-find chemical compounds from animal sources.

Innovation Solution

A neural network-based model is trained on source chemical compounds and their flavors/odors to generate embeddings, allowing for the identification of alternative chemical compounds that recreate the sensory properties of target compounds, using a graph neural network and feed forward network to suggest plant-based substitutes for animal-derived ingredients.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Productivity

If manual laboratory processes are used to develop plant-based food alternatives, then ingredient combinations can be tested, but the process is time-consuming and inefficient

Engineering Contradiction:
Improvedevelopment efficiencyVSAvoidtime to develop food formula
Core Design Contradiction:
ProductivityVSLoss of time

Solution Approach 1:

The patent replaces manual mechanical laboratory processes with an artificial intelligence system that uses machine learning models to predict ingredient combinations and their sensory properties. The AI system processes chemical compound data and generates predictions about plant-based alternatives, eliminating the need for time-consuming manual experimentation and testing in laboratories.

Inventive Principle:
Principle #28Mechanics substitution (Replace mechanical system)

2Object-affected harmful factors

If plant-based alternatives are developed to replace animal-based ingredients, then health and environmental benefits are achieved, but matching taste and texture is difficult

Engineering Contradiction:
Improvehealth and environmental impactVSAvoidsensory property matching
Core Design Contradiction:
Object-affected harmful factorsVSManufacturing precision

Solution Approach 1:

The patent uses the AI system to analyze and match multiple parameters of chemical compounds including sensory properties, chemical structure, and functional characteristics. The machine learning model processes these parameters to identify plant-based compounds that replicate the sensory profile of animal-based ingredients, achieving precise matching of taste and texture while maintaining health and environmental benefits.

Inventive Principle:
Principle #35Parameter changes

3Reliability

If expensive or hard-to-find chemical compounds from animal sources are used, then functional properties can be achieved, but cost and availability become issues

Engineering Contradiction:
Improvefunctional property consistencyVSAvoidingredient availability
Core Design Contradiction:
ReliabilityVSEase of manufacture

Solution Approach 1:

The patent employs the AI system to create virtual copies or representations of animal-based chemical compounds by analyzing their molecular structures and sensory properties. The machine learning model then identifies plant-based compounds that serve as substitutes with similar functional properties, making the ingredients more available and easier to manufacture while maintaining consistent functional performance in food products.

Inventive Principle:
Principle #26Copying

Data Source

PatentEP4133426B1Machine learning driven chemical compound replacement technology
Publication Date: 2024.08.21 NOTCO DELAWARE LLC
  • EP4133426B1 patent drawingFigure 1
  • EP4133426B1 patent drawingFigure 2
  • EP4133426B1 patent drawingFigure 3

AI summary

Techniques to suggest alternative chemical compounds that can be used to recreate or mimic a target flavor using artificial intelligence are disclosed. A neural network based model is trained on source chemical compounds and their corresponding flavors and odors. The neural network-based model learns compound embeddings of the source chemical compounds and a target chemical compound of a food item. From the compound embeddings, one or more chemical compounds that are closest to the target chemical compound may be determined by a distance metric. Each suggested chemical compound is an alternative that can be used to recreate functional features of the target chemical compound.