Inspiration

Healthcare is one of the most critical sectors, yet many systems still rely on fragmented communication, long waiting times, and manual record-keeping. During our observation, we realized that doctors often lack quick access to previous patient interactions, and patients struggle with inefficient consultation processes.

This inspired us to build CareXAI, a platform that combines real-time communication and artificial intelligence to make healthcare more efficient, accessible, and intelligent.

What it does

CareXAI is a real-time telehealth platform that connects patients, doctors, and administrators in a single system.

It provides: Real-time chat between doctor and patient Video consultation using WebRTC/Agora AI-powered health risk prediction AI-based conversation summarization Role-based dashboards (Patient, Doctor, Admin)

One of the key features is: After a consultation, the system generates an AI-powered summary of the conversation, which helps doctors understand patient history in future visits.

How we built it

We built CareXAI using a modern full-stack architecture:

Frontend: React.js, TypeScript, Vite Backend: Node.js, Express.js Database: PostgreSQL with Prisma ORM Real-time communication: Socket.IO Video calls: WebRTC / Agora AI integration: Groq API (LLaMA models)

The system follows a client-server model where the frontend interacts with backend APIs, and real-time updates are handled through WebSockets.

Challenges we ran into Implementing real-time communication reliably using Socket.IO Handling video call integration with WebRTC/Agora Designing accurate AI prompts for medical summarization Managing database schema and relationships using Prisma Deployment issues with cloud services and environment variables Accomplishments that we're proud of Successfully built a complete real-time healthcare system Integrated AI summarization for doctor-patient conversations Designed a multi-role dashboard system Achieved end-to-end working flow (booking → consultation → AI insights) Deployed the system using modern cloud platforms What we learned Building scalable full-stack applications Working with real-time systems (Socket.IO) Integrating AI models into real applications Handling deployment and cloud infrastructure Understanding healthcare workflows and system design What's next for CareXAI – AI-Powered Health Insight Engine Mobile application (Android/iOS) Multi-language support Advanced AI models for better predictions Integration with hospital systems (EHR) Enhanced security and compliance features Advanced analytics for doctors and hospitals

CareXAI aims to transform traditional healthcare into a smart, AI-driven ecosystem that improves both patient experience and clinical decision-making.

Built With

Share this project:

Updates

posted an update

Update: Core Features Successfully Implemented

We have successfully implemented an end-to-end AI-powered healthcare platform focused on accessibility, automation, and early risk detection.

AI-Based Risk Prediction

We implemented machine learning models using XGBoost trained on real-world datasets to predict:

  • Diabetes risk
  • Hypertension risk
  • Heart disease risk These models enable early detection and proactive healthcare decisions.

Smart Prescription & Health Records

We implemented:

  • Handwritten prescription processing (digitization pipeline)
  • Secure storage of patient health records using PostgreSQL
  • Downloadable, auto-generated health reports

Automation & Emergency Support

We implemented:

  • Automated medication and appointment reminders
  • Emergency trigger system that:
    • Alerts family members
    • Connects to emergency services (112)

AI Automation Assistant (Multilingual)

We implemented a user-friendly assistant that:

  • Supports multiple languages
  • Helps users book appointments
  • Allows voice/text interaction to access reports and services
  • Simplifies usage for non-technical and rural users

Communication System

We implemented:

  • Secure video and audio consultation
  • Real-time messaging system (similar to WhatsApp) for doctor–patient communication

Security & Admin Verification

We implemented:

  • JWT-based authentication for secure login
  • Admin verification system for both patients and doctors

Our mission is to build an intelligent, inclusive healthcare ecosystem that bridges the gap between AI technology and real-world accessibility.

Log in or sign up for Devpost to join the conversation.