Estimating patient's length of stay in the Emergency Department with an artificial neural network

AMIA Annu Symp Proc. 2005:2005:1155.

Abstract

Predicting a patient's expected length of stay for an Emergency Department encounter is valuable to anticipate impending operational bottlenecks that may lead to diversion. We developed and validated an artificial neural network using data from >16,000 patients using clinical and operational parameters that are commonly available early during an encounter. Performance on the training set predicted length of stay within an average of 2 hours (sigmae2<500), but declined to an average of 7.5 hours (sigmae2<6000) in the validation set. Chief complaint specific trials using the most frequent chief complaints, however, predicted within an average of 3.5 hours (sigmae2 <145), with similar validation.

Publication types

  • Evaluation Study

MeSH terms

  • Adult
  • Emergency Service, Hospital*
  • Feasibility Studies
  • Humans
  • Length of Stay*
  • Neural Networks, Computer*