Getting Started
Complete setup guide, installation instructions, and configuration documentation for CuraSense - your AI-powered medical diagnosis system.
Latest Updates
Discover all the new features and improvements in the latest version
Next.js 16 Frontend
Complete redesign with React 19 and TypeScript
Framer Motion
Smooth, professional animations throughout
Modern UI/UX
Radix UI components with Tailwind CSS 4
Fully Responsive
Works seamlessly on desktop, tablet, and mobile
Dark/Light Mode
Theme switching with next-themes
Report History
Track and manage all your past diagnoses
AI Chat Assistant
Interactive chat for follow-up questions
State Management
Zustand for efficient client-state management
Powerful Capabilities
Multi-Agent AI Workflows
Orchestrated diagnosis using CrewAI and LangGraph for comprehensive analysis
Prescription Analysis
Upload and analyze medical prescriptions and blood test reports with AI
X-Ray & CT Analysis
Vision AI for analyzing X-rays, CT scans, and MRI images accurately
Medicine Comparison
Compare medications, check interactions, and find safer alternatives
RAG System
Semantic search with ChromaDB vector database for intelligent retrieval
Real-time Streaming
Live diagnosis updates with Server-Sent Events for instant feedback
Secure
Session-based data isolation with 15-minute TTL for privacy compliance
Fast Processing
Sub-30 second response time for comprehensive medical analysis
Before You Begin
System Requirements
Node.js 18+
For Next.js frontend
Python 3.10+
For ML backend
Conda
Miniconda or Anaconda
Setup Instructions
Clone Repository
git clone https://github.com/VaibhavK289/curasense-diagnosis.git
cd curasense-diagnosisSetup Next.js Frontend
cd curasense-frontend
npm installSetup Python Backend
ML Service:
conda create -n curasense_env python=3.10 -y
conda activate curasense_env
cd curasense-ml
pip install -r requirements.txtFastAPI Backend:
conda create -n curasense_vision_env python=3.10 -y
conda activate curasense_vision_env
cd ml-fastapi
pip install -r requirements.txtAPI Keys Setup
curasense-frontend/.env.local
NEXT_PUBLIC_API_URL=http://localhost:8001
NEXT_PUBLIC_ML_API_URL=http://localhost:8000curasense-ml/.env
GOOGLE_API_KEY="your_gemini_api_key"
GROQ_API_KEY="your_groq_api_key"
TAVILY_API_KEY="your_tavily_api_key"
CREWAI_TRACING_ENABLED=trueml-fastapi/.env
GOOGLE_API_KEY="your_gemini_api_key"
GROQ_API_KEY="your_groq_api_key"
TAVILY_API_KEY="your_tavily_api_key"
HUGGINGFACE_TOKEN="your_hf_token"Directory Layout
curasense-diagnosis/
├── curasense-frontend/ # 🆕 Next.js 16 Frontend
│ ├── src/
│ │ ├── app/ # App Router pages
│ │ │ ├── diagnosis/ # Diagnosis pages
│ │ │ │ ├── prescription/ # Prescription analysis
│ │ │ │ └── xray/ # X-ray analysis
│ │ │ ├── medicine/ # Medicine comparison
│ │ │ ├── history/ # Report history
│ │ │ ├── settings/ # User settings
│ │ │ └── help/ # Help & documentation
│ │ ├── components/ # React components
│ │ │ ├── ui/ # Shadcn/UI components
│ │ │ ├── layout/ # Header, Sidebar
│ │ │ ├── motion/ # Animation components
│ │ │ └── backgrounds/ # Background effects
│ │ ├── lib/ # Utilities & API client
│ │ └── styles/ # Design tokens
│ ├── package.json
│ └── tailwind.config.ts
│
├── curasense-ml/ # Python ML Frontend (Legacy)
│ ├── frontend/ # HTML/CSS/JS dashboard
│ ├── src/
│ │ ├── crew/
│ │ │ └── agents_and_tasks.py # CrewAI agents
│ │ ├── hugging_face_ner.py # NER extraction
│ │ └── output_pydantic.py # Data models
│ ├── app.py # FastAPI server
│ ├── flow.py # Workflow orchestration
│ └── requirements.txt
│
├── ml-fastapi/ # Backend API
│ ├── config/
│ │ ├── main_graph.py # Main diagnosis workflow
│ │ ├── rag.py # RAG system
│ │ ├── vision_graph.py # Vision analysis
│ │ ├── medical_summarizer_graph.py
│ │ ├── vectordb.py # ChromaDB management
│ │ └── validate_api.py # API validation
│ ├── cron/
│ │ ├── jobs.py # Scheduled tasks
│ │ └── storage.py # Session storage
│ ├── main.py # Backend server
│ └── requirements.txt
│
├── .gitignore
├── README.md
└── start_servers.batStart the Application
Option 1: Run All Services Manually
Terminal 1 - Next.js Frontend:
cd curasense-frontend
npm run devTerminal 2 - ML Backend:
cd curasense-ml
conda activate curasense_env
uvicorn app:app --host 0.0.0.0 --port 8000 --reloadTerminal 3 - FastAPI Backend:
cd ml-fastapi
conda activate curasense_vision_env
uvicorn main:app --host 0.0.0.0 --port 8001 --reloadOption 2: Use Batch Script (Windows)
start_servers.batAccess Points
Next.js Frontend
http://localhost:3000Legacy Dashboard
http://localhost:8000API Documentation
http://localhost:8001/docsBuilt With Modern Tools
Next.js 16
React framework with App Router
React 19
UI library
TypeScript 5
Type safety
Tailwind CSS 4
Styling
Framer Motion
Animations
Radix UI
Accessible components
Zustand
State management
Lucide Icons
Icon library
Ready to Get Started?
Clone the repository, set up your API keys, and start building with CuraSense.
Made with ❤️ by CuraSense Team • MIT License