ESIP Home

Progressive Mining of Remotely Sensed Data for Environmental and Public Health Applications

Project Description

This project is jointly funded by IBM and NASA under the NASA earth science enterprise program. This is a three-year project from April 1, 1998 to March 31, 2001. This project is one of the 24 recipients of the ESIP-II and ESIP-III awards announced on Dec. 2, 1997. This project is joint collaboration among the SPIRE project in the image information system group at IBM T. J. Watson Research Center, School of Public Health at Johns Hopkins University, at Baltimore, MD and Raytheon, STX at Greenbelt, MD.

In this project, we propose to develop a prototype system to explore the effects of environmental factors on public health through the retrieval and analysis of remotely sensed images. The methods of content-based retrieval are applied to spatial and temporal data in order to investigate factors that affect public health risk such as temperature, precipitation, vegetation, and land use patterns. The information obtained from remotely sensed images is then used to validate the risk assessment models for infectious and environmental diseases or hazards such as Hantavirus Pulmonary Syndrome, air pollution, fire ant infestations and malaria. The prototype system is capable of participating in and communicating with a federation of Earth Science Information Partners (ESIP) servers.

Objectives

To establish a new paradigm for investigating risk factors in the public health community which integrates spatial data mining and statistical modeling into a novel content-based search and retrieval technique on remotely-sensed data,

To generate and provide risk assessment maps for public health problems and diseases with much finer temporal and spatial precision and coverage than has been previously possible,

To interact with other ESIP systems to share and discover data that is related to the development and validation of public health risk assessment models.