MedDoc

An AI chatbot designed to provide foods to avoid based on medications, allergies, and dietary restrictions

Comment

Inspiration

Managing food and medication interactions can be overwhelming, especially for individuals with specific allergies or those on multiple prescriptions. We wanted to create a solution that simplifies this process — a chatbot that helps users understand which foods they should avoid based on their medications and allergies. With the growing integration of AI into healthcare, MedDoc aims to bridge the gap between personalized nutrition and safe medication use.

What it does

MedDoc is an AI-powered chatbot that helps users identify potential food-drug interactions based on their specific allergies and medications. The chatbot guides users through a simple conversation flow:

  1. Asks for allergies.
  2. Requests a list of current medications.
  3. Analyzes the input using OpenAI's GPT-3.5-turbo model and a database of food-drug interaction data.
  4. Highlights which foods to avoid to prevent adverse interactions.

How we built it

Frontend: Built with React.js using a conversational chatbot interface. We used Axios for API calls and styled the app with CSS and TailwindCSS. AI Integration: Leveraged OpenAI’s GPT-3.5-turbo API to process user input and provide tailored recommendations. Backend: Utilized Firebase for storing user data securely. Database: Incorporated a CSV dataset of food-drug interactions from an SQL dataset, preprocessed with Python, to enhance the AI's recommendations. APIs: Integrated OpenAI’s API directly into the React app, handling user inputs and generating responses in real-time.

Challenges we ran into

Real-time AI integration: Managing asynchronous API calls while maintaining a smooth chat experience in React required fine-tuning. Firebase integration: Ensuring user data was properly stored and retrieved in Firebase, while also considering privacy and security.

Accomplishments that we're proud of

Solving real world problems. Figuring out reading SQL and making Figma designs for user readability.

What we learned

Firebase & React integration: We deepened our understanding of managing user data and syncing it with a frontend in real-time.

What's next for MedDoc

Enhanced AI Fine-Tuning: We plan to fine-tune the GPT model using the complete food-drug interaction dataset for even more accurate recommendations.

Built With

Share this project:

Updates