FOODIFY

Inspiration

Food labels often contain ingredient names that are unfamiliar or difficult to interpret. For someone managing a food allergy, simply finding an exact allergen keyword may not be enough to understand whether an ingredient is related to their personal allergen.

Our Idea

FOODIFY is a personalized food-label intelligence system that helps users understand ingredient–allergen relationships.

The user creates a personal allergy profile and scans a packaged food ingredient label using a smartphone camera. FOODIFY uses OCR to extract the ingredient text, processes and normalizes ingredient names, and maps ingredients to related allergens through a structured knowledge layer.

For example:

Whey Protein Concentrate → derived from → Milk → matches → User's Milk Allergy Profile

Instead of only highlighting a word, FOODIFY explains why an ingredient is flagged.

How It Works

Scan → OCR → Ingredient Understanding → Relationship Mapping → Personalized Matching → Explanation

The system is designed to distinguish between identified relationships and uncertain information. If an ingredient–allergen relationship cannot be determined confidently, FOODIFY communicates the uncertainty instead of treating it as automatically safe.

Innovation

The key idea is relationship-aware ingredient analysis rather than simple allergen keyword matching.

FOODIFY focuses on:

Ingredient-derived allergen relationships
Synonyms and alternate ingredient names
Personalized allergy profiles
Explainable results
Uncertainty-aware interpretation

Expected Impact

FOODIFY aims to make food-label interpretation faster, clearer, and more personalized for people managing food allergies, helping them make more informed food choices.

This submission represents our Round 1 concept. A fully developed prototype is planned for the subsequent development stage if shortlisted.