README.md

metadata

title: EWAAST - Equitable Wound Assessment & Simulation Tool
emoji: 🩺
colorFrom: blue
colorTo: green
sdk: docker
app_port: 7860

EWAAST: Equitable Wound Assessment & Simulation Training 🩺✨

An AI-powered platform for inclusive wound care education and assessment.

EWAAST (Equitable Wound Assessment & Simulation Tool) bridges the gap in nursing education regarding skin tone diversity. It replaces biased legacy tools (which often rely on "redness" detection) with an AI-driven, Monk Skin Tone (MST) aware framework.

🌟 Key Features

1. Digital Purpose-T Assessment 📝

A modern replacement for the Waterlow/Braden scale that:

• Forces MST Acknowledgement: Nurses must identify the patient's Monk Skin Tone first.
• Adaptive Visuals:
  • Light Skin (MST 1-6): Asks about redness/erythema.
  • Deep Skin (MST 7-10): Hides redness prompts. Instead, guides users to check for temperature, induration (firmness), and purple/blue discoloration.
• AI Validation: Uses Gemini 1.5 Pro to validate images and ensure quality.

2. Student Mode "Flight Simulator" 🎓

A safe space for student nurses to practice without risk.

• Randomized Patient Scenarios: Pulls from a diverse databank of patients with various skin tones and wound types.
• Blind Assessment: Students must diagnose the wound stage, tissue type, and priority action without help.
• AI Preceptor 🧠: Real-time feedback engine powered by Gemini.
  • Example: "Incorrect. For MST 8, you should not rely on redness. Did you palpate for warmth?"
• Equity Scorecard: Tracks accuracy specifically on Deep Skin Tones to highlight and correct bias.

🚀 Tech Stack

• Frontend: React.js (Custom Hooks, CSS Modules)
• Backend: Flask (Python)
• AI Model: Google Gemini 1.5 Pro (via google-genai SDK)
• Deployment: Dockerized on Hugging Face Spaces

🛠️ Setup & Installation

1. Clone the repository
2. Install Dependencies:

   pip install -r requirements.txt
   cd frontend && npm install
   

3. Environment Variables: Create a .env file with:

   GOOGLE_API_KEY=your_gemini_key_here
   

4. Run Locally:

   # Terminal 1 (Backend)
   python ewaast_app.py
   
   # Terminal 2 (Frontend)
   cd frontend && npm start
   

📚 Pedagogical Framework

EWAAST is built on the EPUAP/NPIAP Guidelines, specifically emphasizing the exclusion of "erythema" as a standalone indicator for dark skin tones. It aims to de-bias nursing judgment through repetitive, corrected simulation.

---

Built by Nurse Citizen Developers.