Clinical Documentation System

Offline NLP-Based Nurse Documentation Assistant

The Clinical Documentation System is an offline-first nursing documentation assistant that accepts natural language input, extracts structured clinical information, maintains a chronological patient timeline, and generates clean nursing-style notes.
The system is strictly limited to documentation tasks and does not provide diagnosis or treatment recommendations.

---

Project Structure (Important)

Backend and frontend must be run separately.

Backend → FastAPI (Python)
NOTE THE BACKEND IS IN THE APP FOLDER HERE FOR SINGLE DEPLOY MOVE OUTSIDE TO WORK PROPERLY Frontend → Next.js / React

Do not run backend and frontend in the same terminal.
Each must run in its own terminal window.

---

Requirements

Backend

• Python 3.8+
• pip
• Internet connection required only for first-time model download

Frontend

• Node.js 18+
• npm / yarn / pnpm / bun

---

Backend Setup (FastAPI)

1. Create and activate virtual environment

python -m venv venv
source venv/bin/activate      # macOS / Linux
# venv\Scripts\activate      # Windows

2. Install Python dependencies

pip install fastapi uvicorn transformers torch faiss-cpu numpy

3. Download models (run once)

python download.py

4. Start backend server

python api_server.py

---

Frontend Setup

Start Next.js dev server

npm run dev

---

Running Summary

Terminal 1 → Backend (FastAPI)
Terminal 2 → Frontend (Next.js)

Backend must be running first.
System works offline after first model download.