This Streamlit application provides interactive analytics and insights from a dataset of 55,500 patient records across 10 U.S. hospitals. It features sections on patient demographics, hospital performance, insurance and billing, and trends and forecasting. The dashboard uses Plotly for visualizations and employs machine learning techniques, including K-Prototypes clustering, to uncover patterns in patient data and billing.
The Onyx Data DNA April 2025 Challenge provided this dataset, and I developed the Streamlit app as part of my competition submission.
- Patient Demographics: Analyze age, gender, blood type distributions, and more, offering insights into patient characteristics.
- Hospital Performance: Evaluate hospital metrics such as patient volume, billing, and efficiency, aiding in performance comparisons.
- Insurance & Billing: Examine billing patterns, insurance provider performance, and condition-specific billing with clustering insights for cost optimization.
- Trends & Forecasting: Explore historical trends and forecast future metrics, supporting strategic planning.
- Clone the repository
- Navigate to the project directory
- Install the required dependencies
- Ensure the dataset CSV file 'Healthcare Analysis Dataset.csv' is placed in the correct directory as specified in
utils.py. - Run the Streamlit app: Home.py
Upon running the app, use the sidebar to select time periods and navigate through different pages to explore various aspects of the healthcare data. Each page provides interactive visualizations and insights, with options for period comparisons and detailed analytics.
This project is licensed under the MIT License.