This project is an AI-powered healthcare system that analyzes patient vitals such as Heart Rate, SpO2, and Temperature to detect potential health risks and generate intelligent medical insights.
It combines rule-based risk scoring with a Large Language Model to provide meaningful and structured health analysis.
Early detection of health risks based on vital signs is crucial, but continuous monitoring and expert analysis are not always accessible. Many patients fail to recognize warning signs in time.
This project aims to provide an AI-driven solution that quickly evaluates patient vitals and assists in early risk identification.
This project is inspired by real-world scenarios where patients experiencing symptoms like low oxygen levels, high fever, or abnormal heart rate require immediate attention, but timely medical consultation may not always be available.
The system is designed to act as a preliminary assistant for risk awareness.
- 📊 Risk Score Calculation (Low / Medium / High)
- 🤖 AI-powered medical insights using Gemma
- 🖥️ Interactive UI using Gradio
- ⚡ Fast and simple health risk analysis
- 🔐 Secure token handling using environment variables
- 🚀 Deployment-ready with Hugging Face Spaces
- Python
- PyTorch
- Transformers (Hugging Face)
- Gradio
- User inputs patient vitals (Heart Rate, SpO2, Temperature)
- System calculates a risk score using predefined conditions
- Risk level is classified as Low, Medium, or High
- Input is passed to an LLM for deeper analysis
- AI generates condition insights and medical advice
Input:
- Heart Rate: 120 bpm
- SpO2: 88%
- Temperature: 102°F
Output:
- Risk Score: High
- Condition: Possible infection or respiratory issue
- Advice: Seek immediate medical attention
AI-HealthCare-Risk-Detection/ ├── app.py ├── AI_Healthcare_Risk_Detection.ipynb ├── requirements.txt ├── README.md ├── ui.png └── output.png
- Integration with real-time health monitoring devices
- Enhanced accuracy using medical datasets
- Mobile application version
- Multi-patient dashboard system
- Emergency alert integration
This project is developed for educational and hackathon purposes only. It is not a substitute for professional medical advice, diagnosis, or treatment.
Built as part of a hackathon project using modern AI tools and frameworks.