Physics-Informed Graph Learning for Spatially Contiguous and Capacity-Constrained Hospital

A 14.9 MB dataset from figshare, last updated on May 11, 2026, by author Lingbo Liu. It includes shapefiles, adjacency lists, and inpatient flow data for Florida, along with results from multiple graph-based clustering algorithms for hospital service area delineation.

Use Cases

• Delineating hospital service areas based on inpatient flow data.
• Comparing spatial clustering algorithms like SGCN-MST and Region2Vec.
• Analyzing spatial adjacency relationships between ZIP Code Tabulation Areas (ZCTAs).
• Reproducing graph-based community detection results for healthcare regions.

Strengths

• Includes multiple data types: shapefiles, CSV adjacency lists, and algorithm results.
• Provides clustering outputs from six distinct methods (SGCN-MST, GCN-T-MST, etc.) for comparison.
• Released under a permissive CC-BY-4.0 license.

Limitations

• Description metadata is limited; actual data quality requires manual inspection after download.
• Row count and column-level documentation are absent; field semantics must be inferred.
• Data is specific to Florida, which may limit generalizability.

Provenance

Source

figshare

Collection Method

Likely derived from public health or administrative records for Florida.

Freshness

Last updated 2026-05-11 13:04:55; freshness should be verified.

Geography

Florida, United States (at the ZCTA level).