Development of machine learning model to predict pandemic and heart disease count

Link to the website: https://jamesjonginbae.wixsite.com/cdc2021healthcare

Predictive Tool on Disease Case Count

This tool aims to support policymakers in implementing health- and socioeconomic-related policies to reduce COVID-19, heart stroke, and coronary heart cases in a selected county.How many lives can we save by improving health insurance and socioeconomic policies in our area? Our tool lets you explore on the latest data on the three diseases, compare with the state average, and predict the case counts with new policies in place

Exploratory Data Analysis

Why is health insurance important?

Our analysis results showed that the areas with lower income and higher unemployment rates are more likely to suffer with higher uninsured rate. Higher uninsured rates were correlated with higher counts of each disease. If you'd like to read more, click below.

Read More

Data Visualization

Tableau dashboard with disease, insurance, and socioeconomic data

Compare COVID-19, heart stroke, coronary heart cases with other states and counties. Each region in the dashboard is presented with disease data and health / socioeconomic factors.

Go to Dashboard Page

Development of machine learning models

XGBoost models on predicting COVID-19, Heart Stroke, and Coronary Heart Cases

How many cases of COVID-19 or heart diseases can we reduce by reducing the uninsured population by 10%? Or by reducing the unemployment rate? These are the questions that policymakers need answered to implement efficient and effective policies. With our tool, you can predict the case counts in your area by changing the parameters like uninsured population, income, unemployment, and more! Click below to try our model!

Try Our Model!

Tools used

• python 3.8 (libraries: xgboost, scikit-learn, pandas, numpy)
• Tableau
• Excel