CarePlanner.AI 🏥

CarePlanner.AI is an intelligent, privacy-focused health management platform designed to empower patients with chronic conditions (such as Diabetes and Hypertension). By leveraging Generative AI and real-time data analytics, it transforms raw health logs into actionable daily plans, insightful visualization, and personalized medical advice.

---

🌟 Key Features

1. 🧠 AI-Powered Daily Planner

The core of CarePlanner.AI is its ability to generate dynamic daily schedules.

• Context-Aware: Analyzes the patient's age, condition, medications, and last 7 days of vital logs.
• Structured Recommendations: Creates a tailored JSON-based plan including:
  • Tasks: Specific times for medication, meals, and exercise.
  • Targets: Calculated safe ranges for Glucose and BP.
  • Insights: AI-driven analysis of recent trends (e.g., "Glucose stability improved by 10%").
• Safety First: Automatically detects critical vital signs and escalates warnings (e.g., "Seek medical attention").

2. 🤖 Dr. AI Assistant (Chatbot) realtime streaming

A highly responsive, context-aware chatbot for instant health queries.

• RAG-Lite Architecture: Injects the patient's profile and recent health logs directly into the prompt context.
• Streaming Responses: Uses Vercel AI SDK patterns and Genkit streaming to deliver "typing-effect" responses for better UX.
• Optimized Prompts: engineered for concise, medically relevant, and empathetic answers.

3. 📊 Real-Time Health Analytics

Visualizes complex health data into easy-to-understand trends.

• Interactive Charts: Uses Recharts to display Glucose, Blood Pressure (Variable Width), and Weight trends over time.
• Trend Analysis: Calculates "Time in Range" (TIR) and visualizes improved/declining health streaks.

4. 📚 Intelligent Knowledge Feed

Delivers curated educational content to keep patients informed.

• Condition-Specific: Fetches articles relevant only to the user's diagnosed condition (e.g., "Low Sodium Diet" for Hypertension).
• AI Summarization: Generates bite-sized summaries and tags for quick reading.

5. 🛡️ Offline-First Architecture

Built for reliability in all network conditions.

• Firestore Persistence: Enables persistentLocalCache to store data on the device.
• Offline Banner: A reactive UI component (OfflineBanner.tsx) that acts as a circuit breaker, notifying users when connectivity is lost while keeping the app functional.

6. ♿ Inclusive Design

• Accessibility (A11y): Fully navigable via keyboard; ARIA labels on all interactive elements.
• Responsive UI: Mobile-first design using Tailwind CSS for a seamless experience on phones and desktops.

---

🛠️ Technology Stack

Category	Technologies
Frontend	Next.js 15 (App Router), React 19, TypeScript
Styling	Tailwind CSS (Dark Mode, Typography)
Icons	Lucide React
Database	Firebase Firestore (NoSQL, Realtime)
AI Layer	Google Genkit, Ollama (Local LLMs)
Validation	Zod (Schema Validation for AI Output)
Charts	Recharts

---

📂 Project Structure

├── app/
│   ├── actions.ts       # Server Actions for AI generation (Plans, Facts, Articles)
│   ├── api/chat/        # Route Handler for Streaming Chatbot Response
│   ├── learn/           # Educational Content Page
│   ├── layout.tsx       # Root Layout (Navbar, Offline Banner, Fonts)
│   └── page.tsx         # Dashboard (Analytics, Daily Plan)
├── components/
│   ├── AnalyticsCards.tsx # Metric Summary Cards
│   ├── ChatWidget.tsx   # Floating AI Chat Interface
│   ├── DailyPlan.tsx    # AI Plan UI Renderer
│   ├── Navbar.tsx       # Responsive Navigation
│   ├── OfflineBanner.tsx# Connectivity Status Indicator
│   └── Trends.tsx       # Recharts Visualization Component
├── lib/
│   ├── ai.ts            # Genkit / Ollama Configuration
│   ├── firebase.ts      # Firebase Init & Persistence Config
│   └── models.ts        # TypeScript Interfaces (Patient, LogEntry)
└── scripts/
    ├── seed_firestore.mjs # Database Seeding Utility

---

🚀 Getting Started

Prerequisites

• Node.js (v18 or higher)
• Ollama running locally (default: http://localhost:11434)
• Firebase Project with Firestore enabled

Installation

1. Clone the Repository

   git clone https://github.com/Yuvaraja28/CarePlanner.AI.git
   cd CarePlanner.AI

2. Install Dependencies

   npm install

3. Environment Configuration Create a .env.local file in the root:

   # Firebase Configuration
   NEXT_PUBLIC_FIREBASE_API_KEY=your_api_key
   NEXT_PUBLIC_FIREBASE_AUTH_DOMAIN=your_project.firebaseapp.com
   NEXT_PUBLIC_FIREBASE_PROJECT_ID=your_project_id
   NEXT_PUBLIC_FIREBASE_STORAGE_BUCKET=your_project.appspot.com
   NEXT_PUBLIC_FIREBASE_MESSAGING_SENDER_ID=your_sender_id
   NEXT_PUBLIC_FIREBASE_APP_ID=your_app_id
   
   # AI Backend
   OLLAMA_URL=http://localhost:11434

4. Seed the Database (Optional for Demo) Populate Firestore with sample patient data:

   node scripts/seed_firestore.mjs

5. Run Development Server

   npm run dev

   Open http://localhost:3000 to view the app.

---

🔮 Future Roadmap

• Wearable Integration: Sync with Apple Health / Google Fit.
• Multi-Profile Support: Clinical dashboard for doctors.
• Voice Interface: Voice-to-text logging for elderly patients.
• Medication Reminders: Push notifications for pill intake.

---

📄 License

This project is licensed under the MIT License - see the LICENSE file for details.