Fig 1.
Influenza-like Illness (ILI) Peak Infections: State-level peak weekly ILI cases in 2022, expressed as a percentage of the total population.
The maps were generated using ArcGIS Pro using shape files from the U.S. Census Bureau (https://www.census.gov/geographies/mapping-files/time-series/geo/carto-boundary-file.html).
Fig 2.
U.S. ILI Vulnerability Map: State-wise vulnerability maps across the United States, highlighting regions with varying levels of socio-economic vulnerability to Influenza-like Illness (ILI).
States are classified as follows (normalized indices in parentheses): Very Low (0.0–0.13), Low (0.13-0.21), Medium (0.21-0.28), High (0.28-0.35), and Very High (0.35–1.0). The maps were generated in ArcGIS Pro using shape files from the U.S. Census Bureau (https://www.census.gov/geographies/mapping-files/time-series/geo/carto-boundary-file.html).
Fig 3.
Indicators for vulnerable states: This figure presents radar plots to compare the normalized values of the 22 indicators across three states very highly vulnerable to Influenza-like Illness (ILI): (a) District of Columbia (DC) (urban), (b) Michigan (MI) (mixed), and (c) New Mexico (NM) (rural).
The value on each section represents the indicator’s normalized score (0 to 1). The Purple segments indicate factors that increase vulnerability (positive directionality), and orange segments indicate factors that decrease vulnerability (negative directionality).
Fig 4.
State-level indicator trends and vulnerability.
This scatter plot displays normalized socio-economic and health indicators (0 to 1) across U.S. states. The size of each data point (bubble) corresponds to the normalized value of the indicator , while the color represents the state’s vulnerability class. The x-axis presents the indicators in descending order of RFR-derived feature importance or weight
, ranging from “Moved from abroad (%)” (highest importance) to “No computer” (lowest importance).
Fig 5.
Vulnerability Index Framework: Schematic representation of the methodology for computing ILI vulnerability index using a Random Forest regression method.