#SupplyDemand

Description

Source code and data for an application (https://rush-covid19.herokuapp.com/) to aid US hospitals in planning their response to the ongoing COVID-19 pandemic. The application forecasts hospital visits, admits, discharges, and needs for hospital beds, ventilators, and personal protective equipment by coupling COVID-19 predictions to models of time lags, patient carry-over, and length-of-stay. Users can choose from seven COVID-19 models, customize 23 parameters, examine trends in testing and hospitalization, and download forecast data. The application accurately predicts the spread of COVID-19 across states and territories and its hospital-level forecasts are in continuous use by our home institution. The application is versatile, easy-to-use, and can help hospitals plan their response to the changing dynamics of COVID-19, while providing a platform for deeper study.

Requirements

• Python 3.12 or greater
• werkzeug==3.0.6
• dash==3.0.1
• gunicorn==23.0.0
• numpy==2.2.4
• pandas==2.2.3
• scipy==1.15.2
• Flask==3.0.3
• plotly==6.0.1
• DateTime==5.5
• pathlib==1.0.1
• dash-core-components==2.0.0
• dash-html-components==2.0.0

How to run this app

Create a virtual environment for running this app with Python 3. Clone this repository and open your terminal/command prompt in the root folder.

python3.12 -m venv venv_python3_12

In Unix system:

source venv_python3_12/bin/activate

In Windows:

venv_python3_12\Scripts\activate

Install all required packages by running:

pip install -r requirements.txt

Run this app locally with:

python app.py

Directories & Files

DataUpdate
A directory containing python scripts for updating data used by the application. To update data run the following scripts in the following order via your terminal window:

1. get_dataframe_dailyreports.py ...downloads and formats state-level cumulative COVID-19 case data from the John's Hopkins CSSE github repository.
2. get_US_dailyreports_JH_testing.py ...downloads and formats state-level COVID-19 testing data from the John's Hopkins CSSE github repository.
3. TestingData.py ...downloads and formats state-level COVID-19 testing data from The Atlantic's COVID-19 Tracking Project.
4. SEIR-SD_States_Update.py ...Generates updated predictions for the SEIR-SD model for each US state and select territory

DataUpdate/data
A directory containing data files used by the application's app_fxns.py file. The data files are:

1. SEIR-SD_States_Update.txt ...Latest updated version of SEIR-SD model predictions
2. Testing_Dataframe_Most_Recent_Day.pkl ...Latest updated version of state-level COVID-19 testing results for the most recent day on record.
3. Testing_Dataframe.pkl ...Latest updated version of state-level COVID-19 testing results for all days on record.
4. COVID-TESTING-DF ...An intermediate file generated by get_US_dailyreports_JH_testing.py and used by TestingData.py to create the .pkl data files listed above.

ForManuscript:
A directory containing files and other directories needed to recreate figures in the associated manuscript. The files are directories are:

1. model_fxns.py ...a script containing functions specific to the applications models
2. ModelFxns_Figs.py ...a script to generate the general predicted forms of each model; associated with Figure 1 of the associated manuscript.
3. results_dataframe.py ...a script to generate results data for each model's predictive accuracy across time and US states and terroritories. The resulting file model_results_dataframe.pkl is located in the ForManuscript/data directory.
4. ModelPerformance_Figs.py ...a script to generate results figures for each model; associated with Figure 1 of the associated manuscript; uses the model_results_dataframe.pkl file.

app.py
The application's main source code.

app_fxns.py
A python script containing functions needed by app.py.

model_fxns.py
A python script containing functions specific to the applications models

The following are files and directories that are necessary components of the plotly dash application structure

1. venv & env -- a directory containing files and directories specific to the application environment and virtual environment
2. flask -- a directory specific to enabling flask app functionality
3. assets -- a directory of images used by the main application
4. requirements.txt -- software requirements of the app
5. runtime.txt -- python version required or best suited for running the app
6. Procfile -- text file specific to gunicorn; contains a single line (web: gunicorn app:server )

Resources

• Dash