Download PDFOpen PDF in browser

Predicting Patient’s Consultation Length in Emergency Departments with Machine Learning

EasyChair Preprint 2578

2 pagesDate: February 5, 2020

Abstract

Predicting consultation length in Emergency Departments (EDs) is an important step to anticipate upcoming operational bottlenecks that may lead to alterations in the provided service. Faced with numerous challenges, such as increased patient volumes, limited inpatient bed availability, and nurse and physician shortages, EDs encounter frequent capacity limitations, which lead to recurrent ED alteration episodes. The consultation length in ED patients might depend on many factors within and outside the ED. For instance, the patient’s demographics, vital signs, arrival time to the ED, triage (severity) level, among others. If consultation length esti- mates were available early in a patient’s ED encounter, capacity limitations could be identified earlier. Hence, we propose two machine learning models to predict consultation length in EDs and thereby create a decision support tool for resource allocation to improve patient care and to reduce patient’s waiting time and ED overcrowding.

Keyphrases: Artificial Neural Networks, Consultation Length, Random Forests, emergency departments

BibTeX entry
BibTeX does not have the right entry for preprints. This is a hack for producing the correct reference:
@booklet{EasyChair:2578,
  author    = {Pronobesh Roy and María I. Restrepo and Jean-Marc Chauny and Nadia Lahrichi and Louis-Martin Rousseau},
  title     = {Predicting Patient’s Consultation Length in Emergency Departments with Machine Learning},
  howpublished = {EasyChair Preprint 2578},
  year      = {EasyChair, 2020}}
Download PDFOpen PDF in browser