NutriScan: An AI-Powered Mobile Application for Ingredient-Based Food Quality Analysis
The increasing consumption of packaged and processed Food has heightened the need for reliability and accessible tools that can help consumers make informed dietary choices. Although nutritional information is provided on product labels, it is often complex, inconsistent, and difficult for the average consumer to interpret in terms of overall health impact. To address this challenge, this project presents NutriScan, an AI-powered mobile application for ingredient-based food quality analysis, designed to analyze food products and provide actionable health insights in real time. Users can use NutriScan to scan product barcodes which retrieves database information, or they can scan ingredient sections when products are not listed in the database. The system uses OpenAI and Google Gemini API integration to process scanned data from barcodes which results in a 1-10 health rating system with individualized feedback and suggestions. The application helps users make better choices through its features which include showing alternative healthy options for rated products and filtering by diet and allergens and visual ingredient markers and multiple product nutritional comparisons. The system enables community members to add products to the database which enhances both accuracy and product coverage through time.The application uses Flutter for front-end developmentand Flask-based APIs for back-end services and a database system to store product information and user preference data. The system design focuses on user friendly interface development with cross-platform capabilities while testing proved its functionality on various devices. The results show NutriScan makes complicated nutritional information accessible while helping people adopt better eating habits and connecting food labels to real-world health knowledge. The system requires future development to add sustainability metrics and multilingual OCR for ingredient detection and advanced image-based product recognition to support diverse market needs.
Publication Details
- Journal
- Zenodo (CERN European Organization for Nuclear Research)
- Published
- 2025-11-21
- DOI
- https://doi.org/10.5281/zenodo.17675104
- Primary Topic
- Nutritional Studies and Diet
- Type
- article
- Field-Weighted Citation Impact
- 0.00