WellBite

WellBite.ai is an AI-powered assistant that helps patients make informed meal choices based on their medical needs. Let's make health prevelant with relevant, real-time, and personalized information.

Comment

💡 Inspiration

We were really enthusiastic the Patient Safety and AllHealth tracks, we wanted to harness that energy into building something worth it!

🍽️ What it does

WellBite.ai is an online AI assistant designed to help patients make food choices that doesn’t harm their health, considering their medical dietary restrictions. A patient can take a picture of their meal, and using AI, the ingredients are analyzed along with portion sizes and nutrition facts. Based on the patient’s profile, it offers instantaneous feedback on whether the meal complies with their dietary restrictions, helping to avert health risks. In addition, doctors have the capability to access and modify patients' dietary restrictions from an intuitive dashboard, ensuring adherence to the diet in real time and improving overall health outcomes.

🛠️ How we built it

We developed WellBite.ai using:

  • React.js for a responsive and user-friendly front end, allowing smooth interactions, such as image uploads.
  • Python and Flask for efficient back-end request handling and AI model integration.
  • MongoDB for secure storage of user data.
  • Cloudflare for fast, secure hosting with scalability.
  • The Llama AI model, combined with Hugging Face’s BERT FDA nutrition model and Clarifai’s food recognition API for accurate analysis of meal ingredients and dietary adherence.

🚧 Challenges we ran into

One major challenge was updating MongoDB collections from the frontend which took considerable time to debug. We also faced challenges with securing user account management and ensuring the AI could accurately analyze a wide variety of foods.

🎉 Accomplishments that we're proud of

We’re proud to have created a functioning app where patients, doctors, and AI interact in real time. We’re especially proud of integrating advanced AI for meal analysis, securely handling patient data, and creating an intuitive interface for users.

📚 What we learned

We gained valuable experience in:

  • MongoDB management,
  • Cloudflare security and hosting,
  • Llama AI image recognition,
  • Applying AI to real-world health scenarios while balancing technical challenges with ease of use.

🔮 What's next for WellBite

We aim to further improve WellBite.ai for official deployment in hospitals and health centers. Our future goals include:

  • Enhancing AI accuracy in food recognition,
  • Expanding the user base, and
  • Meeting healthcare regulations.
Share this project:

Updates