PhysioBook is a comprehensive physiotherapy booking and management platform powered by AI. It features intelligent patient triage, smart therapist matching, outcome tracking, and real-time exercise guidance using computer vision.

• 🔐 Authentication via Clerk (Google, GitHub, Email & Password)
• 📅 Appointment Booking System with 3-step flow
• 📩 Email Notifications for bookings (Resend)
• 📊 Admin Dashboard for managing appointments
• 💳 Subscription Payments with Clerk (Free + 2 Paid Plans)
• 📂 PostgreSQL for data persistence
• 🎨 Tailwind CSS + Shadcn for styling
• ⚡ TanStack Query for data fetching

• Patients describe symptoms in natural language
• AI classifies condition type (acute, chronic, post-op, sports injury, etc.)
• Determines urgency level (emergency to preventive)
• Recommends appropriate specialist
• Identifies red flags requiring immediate attention
• Confidence scoring for transparency

• Matches based on patient triage data
• Considers therapist specialties & certifications
• Factors in availability & location preferences
• Uses historical success rates & patient feedback
• AI-generated personalized recommendations

• Per-patient recovery dashboard with visual charts
• Pain level, mobility, and function score tracking
• Treatment plan & session notes management
• AI-powered progress analysis
• Predicted recovery timeline
• Exercise compliance monitoring

• Real-time pose detection using TensorFlow.js
• Instant form feedback and corrections
• Rep counting with phase detection
• Form scoring and accuracy metrics
• Voice feedback for hands-free exercise
• Support for multiple exercise types

• ML-based prediction of no-show probability
• Risk-based reminder scheduling (SMS/Email)
• Overbooking strategy suggestions
• Appointment confirmation tracking
• Historical pattern analysis

• Secure WebRTC video calls
• Virtual session logging
• Screen sharing for exercise demos
• Chat during sessions

• Role-based access control
• Audit logging for sensitive actions
• HIPAA-compliant data handling
• Encrypted sensitive records

• Framework: Next.js 14 (App Router)
• Language: TypeScript
• Styling: Tailwind CSS + Shadcn/UI
• Authentication: Clerk
• Database: PostgreSQL + Prisma ORM
• AI/ML:
  • OpenAI GPT-4 (Triage & Analysis)
  • TensorFlow.js (Pose Detection)
  • Custom ML (No-Show Prediction)
• Email: Resend
• Charts: Recharts
• Data Fetching: TanStack Query
• Video: WebRTC
• Voice AI: Vapi (Pro plans)

physiobook/
├── prisma/
│   ├── schema.prisma      # Database schema
│   └── seed.ts            # Database seeding
├── public/                # Static assets
├── src/
│   ├── app/
│   │   ├── (auth)/        # Auth pages
│   │   ├── (dashboard)/   # Dashboard pages
│   │   ├── api/           # API routes
│   │   ├── globals.css    # Global styles
│   │   ├── layout.tsx     # Root layout
│   │   └── page.tsx       # Landing page
│   ├── components/
│   │   ├── dashboard/     # Dashboard components
│   │   ├── exercise/      # Exercise guidance
│   │   ├── landing/       # Landing page
│   │   ├── progress/      # Progress tracking
│   │   ├── triage/        # Triage assessment
│   │   └── ui/            # UI components
│   ├── lib/
│   │   ├── ai/            # AI services
│   │   │   ├── triage.ts
│   │   │   ├── matching.ts
│   │   │   ├── no-show-prediction.ts
│   │   │   └── pose-detection.ts
│   │   ├── prisma.ts      # Prisma client
│   │   └── utils.ts       # Utilities
│   └── middleware.ts      # Clerk middleware
├── .env.example           # Environment template
├── package.json
├── tailwind.config.ts
└── tsconfig.json

• Node.js 18+
• PostgreSQL database
• Clerk account
• OpenAI API key
• Resend account

1. Clone the repository:

git clone https://github.com/yourusername/physiobook.git
cd physiobook

2. Install dependencies:

npm install

3. Set up environment variables:

cp .env.example .env
# Edit .env with your credentials

4. Set up the database:

npm run db:push
npm run db:seed

5. Run the development server:

npm run dev

6. Open http://localhost:3000

# Clerk Authentication
NEXT_PUBLIC_CLERK_PUBLISHABLE_KEY=
CLERK_SECRET_KEY=
CLERK_WEBHOOK_SECRET=

# Database
DATABASE_URL=

# OpenAI
OPENAI_API_KEY=

# Vapi (Voice AI)
NEXT_PUBLIC_VAPI_ASSISTANT_ID=
NEXT_PUBLIC_VAPI_API_KEY=

# Resend (Email)
RESEND_API_KEY=

# App
ADMIN_EMAIL=
NEXT_PUBLIC_APP_URL=

• POST /api/triage/analyze - Analyze symptoms with AI

• POST /api/therapists/match - Get AI-matched therapists
• GET /api/therapists/match - List all therapists

• POST /api/booking - Create appointment
• GET /api/booking - Get user's appointments

• POST /api/progress - Log progress record
• GET /api/progress - Get patient progress

• HIPAA-compliant data handling
• Role-based access control (Patient, Therapist, Admin)
• Audit logging for all sensitive operations
• Encrypted sensitive fields
• Secure webhook verification

Contributions are welcome! Please read our contributing guidelines before submitting PRs.

This project is licensed under the MIT License.

• Dentwise by Burak Orkmez - Original template inspiration
• TensorFlow.js - Pose detection
• OpenAI - AI triage
• Clerk - Authentication
• Shadcn/UI - UI components

Built with ❤️ for DevDash 2026 Hackathon