CardioGuard AI

A machine learning-powered healthcare platform that predicts heart disease risk in real time, explains the results, and helps doctors make faster, data-driven decisions.

Comment

Inspiration

Heart disease is one of the leading causes of death worldwide, yet many cases can be prevented with early detection. We were inspired to create a system that makes risk prediction simple, fast, and accessible to both patients and doctors. Our goal was to bridge the gap between complex medical data and easy-to-understand insights.

What it does

CardioGuard AI is a machine learning-powered platform that predicts the risk of heart disease based on patient health data. It:

  • Analyzes inputs like age, cholesterol, blood pressure, etc.
  • Provides a real-time risk prediction
  • Outputs a probability score
  • Highlights key contributing factors
  • Includes both patient-side interaction and a doctor dashboard for monitoring

How we built it

We built the system using:

  • Machine Learning: Logistic Regression / Random Forest (scikit-learn)
  • Frontend & Deployment: Streamlit
  • Data Processing: Pandas and NumPy
  • Visualization: Matplotlib

The architecture was designed to be modular:

  • A thin ML layer for prediction
  • A Streamlit-based UI for user interaction
  • Separate sections for prediction, analytics, and doctor view

Challenges we ran into

  • Cleaning and restructuring messy, duplicated code into a working system
  • Ensuring correct input formatting for the ML model
  • Handling missing datasets and edge cases
  • Balancing simplicity with functionality under time constraints
  • Designing a UI that is both intuitive and informative

Accomplishments that we're proud of

  • Successfully building a complete end-to-end ML application
  • Creating a system usable by both patients and doctors
  • Designing a clean and functional UI for real-time interaction
  • Deploying a working model that provides instant predictions
  • Maintaining a modular and scalable code structure

What we learned

  • How to integrate machine learning into a real-world application
  • The importance of clean architecture and modular design
  • Debugging and handling real-world edge cases
  • Building user-friendly interfaces for technical systems
  • Rapid development and iteration during a hackathon

What's next for CardioGuard AI

  • Integrating real clinical datasets for improved accuracy
  • Adding advanced models and better evaluation metrics
  • Enhancing explainability using techniques like SHAP
  • Expanding to include other health risk predictions
  • Deploying as a full-scale healthcare support platformardioGuard AI

Built With

Share this project:

Updates