Spatial Risk Modeling for Infectious Disease Surveillance Using Population Movement Data

Oct 20, 2018 NIPS 2018 Workshop Spatiotemporal Blind Submission readers: everyone
  • Abstract: We compare three recently proposed subset scan methods for spatial disease surveillance that use movement data from case and control individuals, rather than a single location per individual, in order to identify areas with a high relative risk of infection. We illustrate the use of these methods to detect spatial clusters of dengue infection risk using geo-located data from Twitter classified into infected cases and non-infected controls.
  • Keywords: event detection, risk modeling, spatial scan statistics, mobility patterns, movement data
  • TL;DR: We compare three recently proposed subset scan methods for spatial disease surveillance that use movement data from case and control individuals.
0 Replies

Loading