This Article 
   
 Share 
   
 Bibliographic References 
   
 Add to: 
 
Digg
Furl
Spurl
Blink
Simpy
Google
Del.icio.us
Y!MyWeb
 
 Search 
   
36th Annual Hawaii International Conference on System Sciences (HICSS'03) - Track 6
Big Island, Hawaii
January 06-January 09
ISBN: 0-7695-1874-5
Margaret R. Kraft, VA Hines Hospital
Kevin C. Desouza, University of Illinois at Chicago
Ida Androwich, Loyola University
In the following paper the process of knowledge generation from the Veterans Administration healthcare information system is explored. This inquiry is concerned with predicting length of stay of a subset of the total patient population, specifically those with spinal cord injuries (SCI). Although SCI patients do not present large numbers, they are outliers in the healthcare system due to extended hospital stays and high costs for treatment. Predicting length of stay can increase efficiencies and effectiveness in resource allocation thus lowering cost. The following research is the first of its kind to use nursing diagnosis and neural networks to predict length of stay. Background material on SCI and the knowledge discovery process is introduced. The entire data mining process is described beginning with data gathering followed by cleaning, aggregation, and integration. Issues faced while conducting the research are discussed. Results of artificial neural networks used to predict length of stay are presented.
Citation:
Margaret R. Kraft, Kevin C. Desouza, Ida Androwich, "Data Mining in Healthcare Information Systems: Case Study of a Veterans? Administration Spinal Cord Injury Population," hicss, vol. 6, pp.159a, 36th Annual Hawaii International Conference on System Sciences (HICSS'03) - Track 6, 2003
Usage of this product signifies your acceptance of the Terms of Use.