Project library
COVID-19 Outbreak Drivers
Explore early warning signals for ICU strain and the factors behind them.
STAR Summary
- Situation
- Using 2020-2023 HHS hospital-capacity data, I built an early-warning model to flag states at risk of crossing 90% ICU utilization in the next 7 days.
- Task
- Develop an early-warning model that flags states at risk of exceeding 90% ICU utilization within seven days and explains the main risk drivers.
- Action
-
- Cleaned and enriched 50k+ rows from the HHS hospital-capacity time series; added rolling stats, trends, and 1/3/7/14-day lag features.
- Trained an XGBoost classifier with class-imbalance weighting and a strict time-based train/test split.
- Result
-
- Used SHAP to highlight the top drivers and embedded an interactive plot in the report.
- Top driver was the share of ICU beds occupied by COVID patients.
- Exported daily, per-state risk scores (probability of breaching 90% ICU utilization within 7 days).
COVID Outbreak Drivers
Explore historical ICU utilization and an experimental risk score by state and date.
Open full demo
Start with a state's historical ICU utilization and exploratory risk score, then explore the trend and map.
- Explore the bundled historical risk scores; the dashboard runs in your browser.
- Drag the date slider to move through time.
- Choose a state or territory from the menu, or select a state on the map.
- Expand Risk drivers and hotspots for supporting detail.
