OneStopMed 🏥

AI-Powered Clinical Workbench & Prescription Generator for Indian Medical Practitioners

Features · Tech Stack · Setup · Deploy · Roadmap

---

Overview

OneStopMed is a full-stack medical application built to modernize the prescription workflow for doctors in India. It combines a lightning-fast drug search engine (powered by SQLite FTS5 across 200,000+ medicines), a medical-grade PDF generator, and a secure doctor authentication system — all wrapped in a clean, modern React interface.

Doctors can search drugs by brand or generic name, build prescriptions with smart auto-complete, save reusable templates, and generate professionally formatted PDF prescriptions — ready to print and sign.

---

🚀 Key Features

⚡ Instant Drug Search

• Full-text search across 200,000+ Indian medicines using SQLite FTS5 with Porter stemming.
• Handles partial matches and typos (e.g., "dolo" → "DOLO 650 Tablet").
• Returns brand name, generic composition, drug type, and manufacturer — in milliseconds.

📄 Medical-Grade PDF Generation

• Generates high-resolution, vector-based PDF prescriptions using WeasyPrint + Jinja2 templates.
• Auto-populates doctor credentials (name, qualification, NMC Reg. No., clinic details).
• Includes patient vitals (age, weight, BP, allergies), diagnosis, Rx table, and follow-up instructions.
• Clean, print-ready layout with proper medical formatting (℞ symbol, structured fields).

💊 Prescription Templates

• Save frequently used prescription sets as reusable templates.
• Instantly load, rename, or delete saved templates.
• Templates are doctor-specific and stored securely in Supabase.

🔐 Secure Doctor Authentication

• Supabase Auth with JWT-based session management.
• Registration flow with admin approval (pending registrations table).
• Passwords hashed with bcrypt before storage.
• Protected API routes — every endpoint validates the JWT token server-side.

👤 Doctor Profile Management

• Editable profile with fields: full name, qualification, clinic name, clinic address, phone, NMC registration number.
• Profile data is automatically embedded into generated prescriptions.

🧠 Smart Auto-Complete

• Real-time drug suggestions as you type — with keyboard navigation (↑/↓/Enter/Escape).
• Auto-fills generic composition and drug type on selection.
• Prevents prescription errors by validating against the curated database.

📋 Prescription History

• Every generated prescription is logged to Supabase with full patient + medicine data.
• Enables future analytics and audit trail capabilities.

---

🛠️ Tech Stack

Layer	Technology	Purpose
Frontend	React 19, TypeScript, Vite	SPA with modern tooling
Styling	Tailwind CSS, Shadcn/ui (Radix)	Clean, accessible UI components
Routing	React Router v7	Client-side navigation
Auth (Client)	Supabase JS SDK	Session management, JWT tokens
HTTP Client	Axios	API communication
Backend	FastAPI (Python)	High-performance async API
Server	Uvicorn (ASGI)	Production-grade ASGI server
Drug Search	SQLite FTS5	Full-text search with Porter stemming
PDF Engine	WeasyPrint + Jinja2	CSS-styled PDF rendering
Auth (Server)	Supabase Admin + bcrypt	JWT validation, password hashing
Database	Supabase (PostgreSQL)	Doctors, templates, prescriptions
Drug Data	SQLite (42 MB, FTS5)	200k+ medicines from CSV ingestion
Icons	Lucide React	Consistent iconography

---

📂 Project Structure

OneStopMed_Prescription/
├── backend/                      # Python / FastAPI
│   ├── main.py                   # All API endpoints
│   ├── ingest.py                 # CSV → SQLite FTS5 data pipeline
│   ├── requirements.txt          # Python dependencies
│   ├── data/
│   │   ├── medicine_data.csv     # Raw drug dataset (~27 MB)
│   │   └── drugs.db              # SQLite FTS5 database (~42 MB)
│   └── templates/
│       └── prescription.html     # Jinja2 template for PDF layout
│
├── frontend/                     # React / Vite / TypeScript
│   ├── src/
│   │   ├── App.tsx               # Root component, routing, auth guard
│   │   ├── Login.tsx             # Login page
│   │   ├── Register.tsx          # Registration page
│   │   ├── supabaseClient.ts     # Supabase client initialization
│   │   ├── pages/
│   │   │   ├── Dashboard.tsx     # Main prescription builder UI
│   │   │   └── Profile.tsx       # Doctor profile editor
│   │   └── components/ui/        # Shadcn/ui components (12 files)
│   │       ├── Header.tsx        # App header with navigation
│   │       ├── button.tsx, card.tsx, dialog.tsx, input.tsx,
│   │       │   label.tsx, table.tsx, tabs.tsx, textarea.tsx,
│   │       │   badge.tsx, scroll-area.tsx, separator.tsx
│   │       └── ...
│   ├── vite.config.ts
│   ├── tailwind.config.js
│   └── package.json
│
├── my-app/                       # Next.js (Patient-facing / Landing)
│   ├── app/
│   │   ├── page.tsx              # Landing page
│   │   └── layout.tsx            # Root layout
│   └── package.json
│
├── .gitignore
└── README.md

---

⚡ Getting Started

Prerequisites

Tool	Version	Notes
Python	3.11+	Backend runtime
Node.js	18+	Frontend tooling
GTK3 Runtime	Latest	Required by WeasyPrint on Windows (install guide)
Supabase Project	—	Create one at supabase.com

1. Clone the Repository

git clone https://github.com/your-username/OneStopMed_Prescription.git
cd OneStopMed_Prescription

2. Backend Setup

# Create and activate a virtual environment
python -m venv venv
venv\Scripts\activate          # Windows
# source venv/bin/activate     # macOS/Linux

# Install dependencies
pip install -r backend/requirements.txt

Create backend/.env with your Supabase credentials:

SUPABASE_URL=https://your-project.supabase.co
SUPABASE_KEY=your-service-role-key
CORS_ORIGINS=http://localhost:5173

(Optional) Rebuild the drug database from the raw CSV:

cd backend
python ingest.py

Start the API server:

uvicorn backend.main:app --reload

🟢 API is now live at http://127.0.0.1:8000 — interactive docs at /docs

3. Frontend Setup

cd frontend
npm install

Create frontend/.env.local:

VITE_API_BASE_URL=http://localhost:8000
VITE_SUPABASE_URL=https://your-project.supabase.co
VITE_SUPABASE_ANON_KEY=your-anon-key

Start the dev server:

npm run dev

🟢 Frontend is now live at http://localhost:5173

---

☁️ Deployment

Backend → Render

1. Push your code to GitHub.
2. Create a new Web Service on Render.
3. Connect the repository and set the Root Directory to backend.
4. Render automatically installs Python dependencies and system packages (pango, cairo, gdk-pixbuf) for PDF generation.
5. Add your environment variables (SUPABASE_URL, SUPABASE_KEY, CORS_ORIGINS) in the Render dashboard.

Frontend → Vercel

1. Import the repository on Vercel.
2. Set the Root Directory to frontend.
3. Vercel auto-detects Vite and deploys.
4. Add environment variables (VITE_API_BASE_URL, VITE_SUPABASE_URL, VITE_SUPABASE_ANON_KEY) in the Vercel dashboard.

Important: Set VITE_API_BASE_URL to your Render backend URL (e.g., https://your-app.onrender.com).

---

🔌 API Reference

Method	Endpoint	Auth	Description
POST	/register	✗	Submit a doctor registration application
GET	/me	✓	Get current doctor's profile
PUT	/profile	✓	Update doctor profile
GET	/search?q=	✗	Search drugs by name (min 2 chars)
POST	/generate_pdf	✓	Generate a prescription PDF
POST	/templates	✓	Create a new prescription template
GET	/templates	✓	List all templates for the current doctor
PATCH	/templates/{id}	✓	Rename a template
DELETE	/templates/{id}	✓	Delete a template

---

🗄️ Database Schema (Supabase)

The application uses the following tables in Supabase (PostgreSQL):

Table	Description
pending_registrations	Stores doctor registration requests awaiting admin approval
doctors	Approved doctor profiles (name, qualification, clinic info, NMC reg.)
prescription_templates	Saved prescription templates per doctor
prescriptions	Full prescription history with patient data and medicines

The drug search database (drugs.db) is a local SQLite FTS5 database — not stored in Supabase — for maximum search performance.

---

🗺️ Roadmap

• Phase 1 — Fuzzy / Full-Text Drug Search API
• Phase 2 — React Frontend & Prescription Builder UI
• Phase 3 — PDF Generation Engine
• Phase 4 — Cloud Deployment (Render + Vercel)
• Phase 5 — Doctor Authentication & Login (Supabase Auth)
• Phase 6 — Prescription History & Templates
• Phase 7 — Patient-Facing Portal (Next.js)
• Phase 8 — Dashboard Analytics (Appointment trends, patient footfall)
• Phase 9 — WhatsApp Prescription Sharing Integration

---

🛡️ License & Disclaimer

License: Private / Proprietary

⚕️ Medical Disclaimer: OneStopMed is an assistive technology designed for registered medical practitioners. It does not replace professional medical judgment. All generated prescriptions must be physically reviewed, signed, and verified by a licensed doctor before being issued to patients.

---

🔒 Security Notes

• All API keys and secrets are loaded from environment variables (.env files) — never hardcoded.
• .env files are excluded from version control via .gitignore.
• Supabase service role keys are used server-side only and never exposed to the client.
• CORS origins are configurable via environment variables.
• JWT tokens are validated on every protected endpoint.

---

Built by Tejas D. Jaiprakash
AIML Student & Full-Stack Developer