Eatelligence

Inspiration

Our inspiration comes from personal experience. As people who are on a journey to become healthier, we understand how difficult it is to make the right food choices, especially when you're trying to balance your health with a busy lifestyle. One of our team members losing someone dear to cancer made us realize the importance of preventative health and nutrition in managing and reducing health risks like cancer, heart disease, and diabetes. This motivated us to build Eatelligence—an app that empowers individuals to make healthier food choices and live better lives, regardless of their budget or lifestyle.

What it does

Eatelligence is an AI-enhanced web app that helps users make informed food choices while grocery shopping. By simply scanning a barcode, the app provides nutritional insights, such as whether the item is healthy or not using a health score, and gives healthier alternatives for the user based on the health score of the product. It also has the feature to find common stores that have healthier alternatives to the scanned product, and it displays them so the user can find a product they like and visit their nearby store to buy it.

Our dataset/database

We used this website for our dataset, which has a vast amount of nutritional facts and food data based on barcode, and it had an API we used:

link

How we built it

We built Eatelligence using Python as the main programming language for its backend. Python provided us with the flexibility to handle barcode scanning, data processing, and integrating AI-driven food recommendations. For the barcode scanning feature, we used OpenCV and pyzbar to integrate scanning capabilities. Our choice of Python allowed for rapid development and made it easier for the team to iterate quickly. We also used Google Gemini's API to find the common stores that have alternatives for the specific product that has been scanned. We used Streamlit to power the app's front end which uses Python code to make a simple front end.

Challenges we ran into

One of the main challenges we faced was determining how each team member could contribute effectively, especially when we had varying levels of expertise. Some team members were unfamiliar with certain aspects of the project, like coding for AI integration or setting up the front-end interface. We worked through this by splitting tasks based on each person's strengths and interests. We also supported each other by learning together, and this collaborative effort allowed us to overcome initial uncertainties and make significant progress.

Accomplishments that we're proud of

Barcode Scanning Function: We successfully integrated a barcode scanning feature that allows users to scan food items and receive immediate information about their healthiness and price. AI Integration: Our AI-enhanced recommendation system is capable of finding common stores with healthier alternatives matching the type of food product scanned. Team Collaboration: Despite the challenges, we effectively worked together, learned new skills, and successfully integrated multiple features within a short time.

What we learned

Coding the Barcode Scanning Function:

One of the key technical learnings involved coding the barcode scanning function. This involved integrating external libraries or APIs to make it possible to scan and process the barcode data, allowing the app to access nutritional information about the scanned items.

Working with APIs for Food Data:

Using external APIs (like Open Food Facts) to pull nutritional data for scanned products was a great learning experience, as it required understanding how to query and display data in real time.

Building AI-Based Recommendations:

Implementing AI for personalized food suggestions based on the scanned product helped the team understand how machine learning algorithms can be applied to real-world applications for health and wellness.

What's next for Eatelligence

Expand Food Database & API Integrations:

Adding more food options, including fresh produce and local brands, will make the app more robust and provide a greater variety of choices. Integrating more food databases and APIs will also improve accuracy and coverage.

User Personalization:

Users can create personalized profiles, specifying their health goals, dietary restrictions, and preferences. The AI then tailors food recommendations to match their individual needs, such as suggesting lower-sugar or heart-healthy options. The app can also include fun customization features like avatars to make the experience more engaging and user-friendly.

Advanced AI for Health Conditions:

The AI could be further developed to recommend food choices tailored to specific health conditions (e.g., diabetes, high blood pressure). This would make the app more useful for people with specific dietary needs. Enhanced User Experience (UX) & Accessibility Features:

Adding voice control, especially for visually impaired users, can make the app more inclusive. Additionally, refining the overall user experience to make it even more intuitive is a priority. Integration with Fitness Apps & Wearables:

Syncing the app with popular fitness trackers like Fitbit or Apple Health to automatically adjust recommendations based on activity levels or health goals would enhance the personalization aspect.