Falls Prediction in Care Homes Using Mobile App Data Collection

Ofir Dvir, Paul Wolfson, Laurence Lovat, Robert Moskovitch

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

5 Scopus citations

Abstract

Falls are one of the leading causes of unintentional injury related deaths in older adults. Although, falls among elderly is a well documented phenomena; falls of care homes’ residents was under-researched, mainly due to the lack of documented data. In this study, we use data from over 1,769 care homes and 68,200 residents across the UK, which is based on carers who routinely documented the residents’ activities, using the Mobile Care Monitoring mobile app over three years. This study focuses on predicting the first fall of elderly living in care homes a week ahead. We intend to predict continuously based on a time window of the last weeks. Due to the intrinsic longitudinal nature of the data and its heterogeneity, we employ the use of Temporal Abstraction and Time Intervals Related Patterns discovery, which are used as features for classification. We had designed an experiment that reflects real-life conditions to evaluate the framework. Using four weeks of observation time window performed best.

Original languageEnglish
Title of host publicationArtificial Intelligence in Medicine - 18th International Conference on Artificial Intelligence in Medicine, AIME 2020, Proceedings
EditorsMartin Michalowski, Robert Moskovitch
PublisherSpringer Science and Business Media Deutschland GmbH
Pages403-413
Number of pages11
ISBN (Print)9783030591366
DOIs
StatePublished - 1 Jan 2020
Event18th International Conference on Artificial Intelligence in Medicine, AIME 2020 - Minneapolis, United States
Duration: 25 Aug 202028 Aug 2020

Publication series

NameLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Volume12299 LNAI
ISSN (Print)0302-9743
ISSN (Electronic)1611-3349

Conference

Conference18th International Conference on Artificial Intelligence in Medicine, AIME 2020
Country/TerritoryUnited States
CityMinneapolis
Period25/08/2028/08/20

Keywords

  • Falls prediction
  • Outcomes prediction
  • Temporal data mining

ASJC Scopus subject areas

  • Theoretical Computer Science
  • General Computer Science

Fingerprint

Dive into the research topics of 'Falls Prediction in Care Homes Using Mobile App Data Collection'. Together they form a unique fingerprint.

Cite this