Professor Ping Yu

Research study uses machine learnings to identify behaviours of aged care residents

A pioneering research study has used artificial intelligence (AI) technologies to understand the behaviours of residents living with dementia in residential aged care facilities (RACF). The study is the first to apply machine learnings to uncover key health and care knowledge from routinely documented, free text nursing notes.  

Professor Ping Yu and her research team from the Illawarra Health and Medical Research Institute (IHMRI) and the Centre for Digital Transformation, University of Wollongong (UOW) led the study. Digital health records from 40 RACFs across New South Wales and Queensland were analysed. The team developed an AI-driven program to accurately process and analyse nursing progressing notes from over 890,000 records.

“We developed a model that allowed for linguistics to be correctly identified across the nursing notes. This data is often unstructured, free text information such as clinical observations that often proven difficult to analyse at such scale,” said Professor Yu.  

Findings from the study uncovered a range of agitated behaviours in residents living with dementia, from physical agitation, verbal disruption, and emotional distress.

“Our model achieved 89% sensitivity, meaning that 89% of people with agitated behaviours were correctly identified.”

This research is hoped to open new pathways for answering many research questions in a relatively short time frame without the need to invest in resource heavy research.

“In our study 4 to 17 behavioural changes were found in half the residents with agitation in dementia. This likely indicates a high risk of care giver burden and the need to redesign the model of care,” said Professor Yu. 

According to the World Health Organisation, Dementia is one of the major causes of disability and dependency among older adults. In Australia, expenditure on dementia care is $15billion and is the second leading cause of death of Australians.

With over 244,000 Australians living in RACF’s and just over half of these residents having dementia, these facilities provide essential care services for older Australians. AI-driven research can play a crucial role in helping to review care needs in RACFs and how these can be addressed.

“In future studies we hope to apply similar methods to identify the relevant nursing actions to address the identified agitated behaviours in dementia and the outcomes,” Professor Yu added.

Media contact

Emily O’Keefe, Marketing and Communications Manager

t: 02 4221 3620



You may also like

Research findings challenge Type 2 diabetes advice
IHMRI revises strategic direction
Karen and Thomas ARC Grants
ARC Future Fellows announced