HG-Medicura-AI

HG-Medicura-AI: Your smart healthcare assistant with AI-powered symptom analysis, drug interaction check, medical term and report summarization.

Comment

Headline

“AI-powered healthcare assistant with symptom analysis, report summarization, and specialist AI agents for smarter medical support.”

Inspiration

Healthcare is often confusing and time-sensitive. Many people struggle to analyze symptoms, understand medical reports, or check drug interactions without visiting a doctor immediately. I wanted to create an AI-powered solution that makes healthcare support more accessible, faster, and user-friendly.

What it does

HG-Medicura-AI is an AI-powered healthcare assistant that helps users:

  • 🔍 Analyze symptoms and get quick insights
  • 💊 Check drug interactions to avoid harmful combinations
  • 📖 Look up medical terms for better understanding
  • 📝 Summarize medical reports into simple language
  • 🚨 Access emergency contacts for quick help
  • 🧭 Triage Agent for prioritizing health concerns

How we built it

  • Frontend: Next.js + React + Tailwind CSS for a clean and responsive UI
  • Backend / APIs: Integrated AI-powered APIs for natural language understanding and analysis
  • Languages: TypeScript and Python for type safety and AI/ML workflows
  • Databases: MongoDB (frontend data handling) + TiDB (distributed backend database)
  • Deployment: GitHub + Vercel (frontend) + Railway (backend)

Challenges we ran into

  • Making AI output accurate yet simple for non-medical users
  • Handling complex medical terminology in a user-friendly way
  • Ensuring the platform is responsive and smooth across devices

Accomplishments that we’re proud of

  • Built a working prototype within limited hackathon time
  • Integrated multiple features (symptom analyzer, drug checker, medical term lookup, report summarizer, triage agent) into one unified assistant
  • Designed a modern, intuitive UI that feels professional and easy to use

What we learned

  • How to integrate AI into real-world healthcare use cases
  • Building scalable frontends with Next.js & TypeScript
  • Developing robust backends with TiDB, OpenAI Agents SDK, Node.js, and Python
  • Managing time and features effectively in a hackathon environment

What’s next for HG-Medicura-AI

  • 🎙️ Adding voice-based support for accessibility
  • 🩺 Partnering with healthcare experts to improve AI accuracy
  • 🔐 Enabling secure user health record storage
  • 📱 Expanding to a mobile application for wider adoption

  • 🧠 Introducing AI Specialist Services (Agents):

    • Cardiology AI – Heart health assessment
    • Dermatology AI – Skin condition analysis
    • Neurology AI – Neurological symptom analysis
    • Pulmonology AI – Respiratory health assessment
    • Ophthalmology AI – Eye health analysis
    • Dental AI – Oral health assessment
    • Allergy & Immunology AI – Allergy and immune system analysis
    • Pediatrics AI – Child health assessment
    • Orthopedics AI – Bone and joint health
    • Mental Health AI – Mental wellness assessment
    • Endocrinology AI – Hormone and metabolic health
    • Gastroenterology AI – Digestive system health
    • Radiology AI – Medical imaging analysis
    • Infectious Disease AI – Infection and disease analysis
    • Vaccination Advisor AI – Personalized vaccination guidance

  • ⚡ Improving Error Handling & UX:

    • Loading states for smooth AI interaction
    • Custom 404 (Not Found) Page
    • Error Boundary Page for handling unexpected issues gracefully

Built With

  • Next.js – Frontend framework
  • React + Hooks + React Icons – UI components & interactivity
  • Tailwind CSS – Styling & responsive design
  • TypeScript – Type safety and scalability
  • OpenAI Agents SDK – AI integration for healthcare analysis
  • MongoDB – Frontend data handling
  • TiDB – Backend database (distributed SQL)
  • Node.js – Server-side logic
  • Python – AI/ML integrations
  • Vercel – Frontend deployment & hosting
  • Railway – Backend deployment & hosting
  • GitHub – Version control

Built With

Share this project:

Updates