GlycoSafe

Introduction

Diabetes management requires careful monitoring of food intake, especially the glycemic load of meals, which directly impacts blood sugar levels. However, most individuals, particularly those in low-resource settings, struggle to manually estimate the glycemic load of their meals. Traditional methods involve referencing food databases, weighing portions, and performing manual calculations, which can be time-consuming and prone to inaccuracies.

Additionally, current dietary assessment tools often rely on subjective user input, leading to potential errors and inconsistencies. There is a need for an automated, image-assisted approach to simplify this process and provide reliable glycemic load estimates.

The Solution

GlycoSafe is an AI-powered mobile application that estimates the glycemic load of meals using image recognition. Users simply capture a photo of their meal, and the system processes the image to identify food items, estimate portion sizes, and calculate the corresponding glycemic load.

Key Features:

• Image-Based Food Recognition: Uses machine learning models to analyze food images and detect meal components.
• Glycemic Load Calculation: Automatically computes the glycemic load based on recognized food items and portion sizes.
• Nutritional Insights: Provides a detailed breakdown of macronutrients and dietary recommendations.
• User-Friendly Interface: Enables diabetic patients and health-conscious individuals to monitor their diet effortlessly.
• Integration with Health Tools: Allows users to log meals, track historical data, and receive personalized insights for better diabetes management.

Technologies Used

Frontend:

• Flutter - For building a cross-platform mobile application.
• Dart - The programming language used for the frontend.
• Firebase - For user authentication and real-time database management.

Backend:

• FastAPI - A high-performance web framework for processing meal images.
• Python - Used for backend logic and AI model implementation.
• TensorFlow/Keras - For training and deploying deep learning models for food recognition.
• PostgreSQL - For storing food data and user logs.
• Docker - For containerizing the backend services.

AI & Image Processing:

• OpenCV - For image preprocessing and analysis.
• YOLO/ResNet - For food recognition and classification.
• NVIDIA Hardware - Used during the prototype development phase for model training.

DevOps & Deployment:

• GitHub Actions - For CI/CD automation.
• Google Cloud/AWS - For cloud hosting and scalable backend deployment.
• GitHub Pages - For project documentation and web-based details.

---

Contributing

If you’d like to contribute to GlycoSafe, please follow the contribution guidelines in CONTRIBUTING.md.

License

GlycoSafe is released under the MIT License. See LICENSE for details.

For more information, visit the Glycosafe Website.