DiagnoAI

AI That Understands, Diagnoses, and Empowers

Comment

Inspiration

Doesn't matter whether getting a doctor’s appointment or standing in an emergency line it can take significant wait times. For appointments it is even weeks, leaving patients without support. During this wait, DiagnoAI acts as a companion, allowing them to voice concerns and receive guidance. Before their appointment, they can generate a detailed diagnosis report, helping doctors better understand their condition for a more informed consultation.

What it does

DiagnoAI is an AI-powered medical chatbot designed to provide friendship. Areal-time health guidance through voice-to-voice interaction and facial emotion recognition. Users can speak their symptoms instead of typing, making it more accessible for those with disabilities or limited mobility. The chatbot also analyzes facial expressions in the background, adapting its responses based on the user’s emotional state. Additionally, DiagnoAI generates detailed diagnosis reports, summarizing symptoms and concerns to help doctors gain insights into the patient's condition over time

How we built it

Frontend: Built using Next.js, Tailwind CSS, and ShadCN ensuring a clean and responsive UI. Backend: Developed with Python and Flask, integrating AI models for speech to speech processing, facial recognition, and diagnosis generation.

Deployed on custom PC using Cloudfare.

Challenges we ran into

Implementing real-time voice-to-voice processing while maintaining accuracy and responsiveness. Facial expression awareness seamlessly building into the chatbot experience without disrupting user interaction. Ensuring efficient diagnosis reports while balancing speed and accuracy.

Accomplishments that we're proud of

Successfully integrated our backend RAG (Retrieval-Augmented Generation) system with our frontend. Implemented a fully functional voice-to-voice feature, making DiagnAi more accessible and user-friendly. Developed emotion-aware AI responses, allowing the chatbot to adapt its tone and guidance based on user emotions.

What we learned

  1. Deployment of Stateless application stack

What's next for DiagnoAI

  • Big Picture: Getting into the hands of people who really need it. True multi lingual support. Finishing the integration of our face awareness functionality into the chat. MEIM audio live broadcasting over HTTPS for ultra fast latency.

Built With

Share this project:

Updates