Online Program

Return to main conference page

All Times EDT

Thursday, June 4
Data Visualization
Education
Education and Data Visualization Posters
Thu, Jun 4, 10:00 AM - 1:00 PM
TBD
 

Building an Open-Sourced Geospatial Visualization Shiny Application in R for Healthcare Providers and Evaluators (308476)

*Dar'ya Y Pozhidayeva, Oregon Health & Science University 
Amy Wilson, Oregon Health & Science University 
Adrienne Zell, Oregon Health & Science University 

Keywords: R, rayshader, leaflet, ggmap, shiny, geospatial data, geocoding, healthcare

We developed an application for geospatial visualization that addresses the lack of easily available open-sourced mapping software. Our solution provides an approachable, reproducible pipeline, which can be used reliably for a number of location-based healthcare datasets. Healthcare is a field that regularly uses maps to illustrate indicators such as accessibility, but which often lacks resources and/or the ability to obtain them quickly. This limitation impacts community and researcher access to software such as Geographic Information System (GIS) which, although reliable and widely utilized, is also expensive and complex. With the goal of increasing the approachability of geospatial visualization, our team selected three distinct location-based datasets representing available healthcare provider and patient locations across Oregon and used them to develop a mapping and visualization application in R and R shiny. Two datasets represent locations of medical and dental providers obtained from state-maintained databases, while a third contains de-identified patient locations generated from a snapshot of Oregon zip codes data pulled from the Marketing Systems Group (MSG) via random, stratified sampling with oversampling for rural counties. The application geocodes input data and analyzes pairwise distances between the nearest available providers and sample patient locations. Distances are then visualized using R packages such as leaflet, rayshader, and ggmap. This poster showcases a step-by-step demonstration of the application in the analysis of a practice data set containing patient and provider locations. The output generates a table of calculated distances between nearest available providers and patient locations which are visualized in a choropleth map. The results generated by this analysis allow researchers using location-based healthcare data to answer questions regarding healthcare access and disparities in healthcare related to geographic location.