Dementia is a word that strikes fear into the hearts of many Australians. With more than 400,000 people currently living with dementia in Australia—and that number expected to skyrocket in the coming decades—early detection and proper care have never been more important.
But what if there were a way to spot dementia earlier, more accurately, and ensure that no one slips through the cracks?
According to the World Alzheimer Report, around 50 million people worldwide are living with dementia, and this figure is expected to triple by 2050.
Here in Australia, dementia is already the second leading cause of death, and the leading cause of death for women. Yet, despite its prevalence, dementia often goes undiagnosed or is picked up too late.
Traditional methods rely on medical coders sifting through mountains of handwritten notes and electronic records, a time—consuming process prone to human error.
This means many people living with dementia may not be counted in official statistics, and worse, may not receive the care and support they desperately need.
Researchers from the National Centre for Healthy Ageing (NCHA)—a partnership between Monash University and Peninsula Health—are working to change this.
Their latest study, supported by national health bodies and published in the prestigious Alzheimer’s & Dementia Journal, has shown that combining traditional data analysis with cutting-edge artificial intelligence (AI) can dramatically improve the accuracy of dementia detection in hospitals.
So, how does it work? The team used a type of AI called natural language processing (NLP), which is designed to ‘read’ and understand the free text in medical records—think doctor’s notes, nursing observations, and discharge summaries.
By analysing structured data (like age, medications, and hospital visits) and the unstructured data (the written notes), the AI can pick up subtle clues that might otherwise be missed.
For example, a nurse might jot down that a patient was ‘confused overnight’ or ‘seemed forgetful during medication rounds’.
While these notes might not trigger a formal diagnosis, the AI can flag them as potential signs of dementia, prompting further investigation.
The study involved over 1,000 people aged 60 and above from the Frankston-Mornington Peninsula region. The AI-powered algorithms could identify people with dementia with a high degree of accuracy, much better than traditional methods alone.
Dr Taya Collyer, the lead author, explained that the research used ‘gold standard’ diagnoses from specialists as a benchmark, ensuring the results were reliable.
The AI didn’t just look for the obvious; it also considered factors like socioeconomic status, medication history, emergency visits, and even in-hospital events like episodes of confusion or distress.
Clinical experts were involved every step of the way to make sure the AI’s findings were relevant and useful in a real-world setting.
Professor Velandai Srikanth, director of the NCHA and project lead, believes this new approach could be a game-changer.
‘Many people are missing out on good care because we are not very good at identifying them or their needs,’ he said.
By harnessing AI, hospitals could spot dementia earlier, ensuring people get the help they need before their condition worsens.
The NCHA’s Healthy Ageing Data Platform is an Australian-first initiative integrating data from electronic health records, safety systems, and other sources.
With the support of the National Health and Medical Research Council, the Medical Research Future Fund, and the Department of Health & Aged Care, this project is paving the way for more innovative, compassionate healthcare.
What do you think about using AI in healthcare, especially for detecting conditions like dementia? Do you have concerns or see it as a step forward? Share your thoughts and experiences in the comments below—we’d love to hear from you!
Also read: The surprising new ally in the fight against dementia
