HeartPredict is a Python library designed to analyze and predict heart failure outcomes using patient data.

• Dataset information
• Key Questions to Answer with the Dataset
  • Descriptive Analysis
  • Correlation and Feature Importance
  • Predictive Analysis
  • Survival Analysis
  • Risk Factor Analysis
• Usage
  • Installation
  • CLI
  • Docker
  • Notebook
• Contributing
• Code of Conduct
• License

The dataset used for this analysis was obtained from kaggle.com. It contains 5000 medical records of patients who had heart-failure and is licensed under CC0; made available under this URL.

• What are the basic statistics (mean, median, standard deviation) of the clinical features?
• How is the age distribution of patients?
• What is the proportion of patients with conditions like anaemia, diabetes and high blood pressure?

• Which clinical features are most strongly correlated with the DEATH_EVENT? And what are the most important features for predicting heart failure outcomes?
• How do different clinical features contribute to the risk of death due to heart failure?

• How accurately can we predict DEATH_EVENT using clinical features?
• Which machine learning model performs best for this prediction task?

• Can we identify patient subgroups with higher or lower survival probabilities?
• And what is the survival rate of patients over the follow-up period?

• How does smoking affect the risk of death in heart failure patients?
• What is the impact of serum creatinine and serum sodium levels on patient outcomes?

You may want to use a virtual environment to install into.

pip install git+https://github.com/HeartPredict/HeartPredict

Once installed, the hp CLI app should be available in your virtual environment or system. You can simply run hp to get a list of available options and commands.

You can also use the CLI via docker by cloning the repository and running the following command:

docker build -t hp --rm . && docker run -it --name hp --rm hp

When you're done, simply exit the container with exit.

We also provide you with an interactive Jupyter Notebook that visualizes our results. It can be found here.

We welcome contributions from the community! If you're interested in contributing to HeartPredict, please take a look at our CONTRIBUTING.md file. It contains all the guidelines you need to follow to get started, including how to report issues, suggest features, and submit code.

We are committed to providing a friendly, safe and welcoming environment for everyone. Please read our Code of Conduct to understand the standards we expect all members of our community to adhere to.

This project is licensed under the MIT License - see the LICENSE file for details.