Predicting asthma exacerbations using artificial intelligence

Joseph Finkelstein, Jeffrey Wood

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

25 Scopus citations

Abstract

Modern telemonitoring systems identify a serious patient deterioration when it already occurred. It would be much more beneficial if the upcoming clinical deterioration were identified ahead of time even before a patient actually experiences it. The goal of this study was to assess artificial intelligence approaches which potentially can be used in telemonitoring systems for advance prediction of changes in disease severity before they actually occur. The study dataset was based on daily self-reports submitted by 26 adult asthma patients during home telemonitoring consisting of 7001 records. Two classification algorithms were employed for building predictive models: naïve Bayesian classifier and support vector machines. Using a 7-day window, a support vector machine was able to predict asthma exacerbation to occur on the day 8 with the accuracy of 0.80, sensitivity of 0.84 and specificity of 0.80. Our study showed that methods of artificial intelligence have significant potential in developing individualized decision support for chronic disease telemonitoring systems.

Original languageEnglish
Title of host publicationInformatics, Management and Technology in Healthcare
PublisherIOS Press
Pages56-58
Number of pages3
ISBN (Print)9781614992752
DOIs
StatePublished - 2013
Externally publishedYes
EventInternational Conference on Informatics, Management, and Technology in Healthcare, ICIMTH 2013 - Athens, Greece
Duration: 5 Jul 20137 Jul 2013

Publication series

NameStudies in Health Technology and Informatics
Volume190
ISSN (Print)0926-9630
ISSN (Electronic)1879-8365

Conference

ConferenceInternational Conference on Informatics, Management, and Technology in Healthcare, ICIMTH 2013
Country/TerritoryGreece
CityAthens
Period5/07/137/07/13

Keywords

  • Computer-assisted instruction
  • health education
  • internet search

Fingerprint

Dive into the research topics of 'Predicting asthma exacerbations using artificial intelligence'. Together they form a unique fingerprint.

Cite this