SymptomSavvy is an AI-powered web application created for Problem Statement A at the Harvard Rare Disease Hackathon 2025. Our platform streamlines early diagnostic workflows for rare diseases by:
- Analyzing patient-reported symptoms
- Matching clinical patterns against rare disease databases
- Surfacing relevant medical literature for verification and next steps
We aim to reduce diagnostic delays and empower both patients and clinicians with actionable insights.
- Symptom pattern recognition with intelligent similarity scoring
- Ranked differential diagnosis suggestions
- Direct access to medical research sources
- Clean and intuitive web interface
- Lightweight deployment for prototyping use
- Backend: Python, Flask
- Frontend: HTML, CSS, JavaScript
- Data Sources: HPO-based rare disease mappings
- Python 3.x
- pip (Python package manager)
- Virtual environment (recommended)
git clone https://github.com/flashyash/savvy
cd savvy
# Set up virtual environment
python -m venv venv
source venv/bin/activate # macOS/Linux
venv\Scripts\activate # Windows
# Install dependencies
pip install -r requirements.txt
# Launch development server
python app.py