The National Centre for Healthy Ageing, a research institute by Monash University and Peninsula Health in Australia, has developed a new AI-driven method to efficiently identify dementia in hospitals.
It researched ways to uplift dementia detection in healthcare settings by combining traditional and AI-driven case identification approaches.
FINDINGS
Based on a media release, the NCHA research team developed dementia-finding algorithms in two streams: a traditional stream for routinely collected structured data from EMRs on the Healthy Ageing Data Platform, and an AI stream for unstructured text records, powered by natural language processing (NLP) and guided by clinical experts.
“Special software was used to harness the large amount of free text data in a way that NLP could then be applied,” added Dr Taya Collyer, one of the research leads.
Besides standard dementia codes, the research also considered information on demographics, socioeconomic status, medications, emergency and clinic health utilisation, and in-hospital events.
The algorithms were tested in a study involving over 1,000 seniors aged 60 and above in Frankston-Mornington Peninsula. It demonstrated “high classification accuracy” – 72.2% specificity and 80.6% sensitivity – in identifying persons with or without dementia, showcasing a potential to improve how dementia is detected, counted, and managed in healthcare settings.
The research received grants from the National Health and Medical Research Council, the Medical Research Future Fund, and the Department of Health and Aged Care.
WHY IT MATTERS
The number of people living with dementia is expected to rise to 150 million worldwide by 2050, according to the World Alzheimer Report. To prepare for this, “[a]ccurate identification is critical to understanding the true size of the problem nationally, and to be able to effectively plan services,” said Monash University.
MARKET SNAPSHOT
While tools for algorithmic detection of dementia are widely available, these often do not use clinically meaningful case definitions and have relied on proxies, such as diagnostic codes or medications, to ascertain cases, claims the NCHA research team. Hence, they took a dual algorithm-based approach in sifting through structured and unstructured EMR data, with one leveraging NLP and its untapped potential.
Large language models (LLMs) have also seen growing application in the detection of cognitive impairment. In super-ageing South Korea, for example, a study by the Electronics and Telecommunications Research Institute recently showed high accuracy of an LLM-driven model in recognising Alzheimer’s disease.
A gamification approach has also demonstrated effectiveness in screening mild cognitive impairment as part of health screening in Singapore.
ON THE RECORD
“Given that clinical recognition of people diagnosed with dementia presenting to hospitals is poor, using this new approach, we could be identifying people earlier for appropriate diagnostic and clinical care. I am sure that many people are missing out on good care because we are not very good at identifying them or their needs,” said NCHA director and project lead Professor Velandai Srikanth.
Credit: Source link
