We may soon have new technology that will detect when someone is about to fall, with the help of AI.
Senior Research Fellow Dr Rena Logothetis is Deakin University’s lead investigator on the project and is working with Home Guardian, a company which designs and markets fall alarm products.
Together they are looking to enhance the effectiveness of its fall detector device using AI technology.
Speaking to Neos Kosmos, Dr Logothetis said falls are a serious issue and are in fact the leading cause of serious injury and injury-related death in Australia.
“Everybody that you speak to has problems with falls,” she says.
“A lot of the work that I do is with industry partners in health – I’ve worked very closely with hospitals as well as aged care facilities and it’s the number one problem.”
While she can’t go into too much information because there is IP involved, the way this technology works is it looks at the activities of an elderly person and determines which of those activities lead to a fall.
Treatment for falls is costing the federal government around $4 billion a year says Dr Logothetis, who also hopes to see the way we deal with the issue become preventative rather than reactive.
“That’s how a lot of the medical field is, once it’s done, let’s go fix it. It is once the fall has happened.”
“The purpose is to try and predict the fall, to be able to intervene prior to a fall occurring.”
She says intervening would involve sending an alert to loved ones or carers and nurses.
“If you’re in an aged care facility, obviously you’ll have the nursing team that would be there and then they would be able to intervene and prevent the fall from occurring.”
When asked by Neos Kosmos how does this technology take into account something like a person randomly tripping over, Dr Logothetis says some stuff may not be picked up but in the future it very well could.
“The thing with randomness, a lot of the times isn’t that random. People fall because they’ve tripped over something.”
“You could possibly train AI models to be able to look at that.”
For those concerned about AI, she says there is still a lot of human involvement from manually labelling and training the AI models to actually building it.
But from the perspective of the user?
“The way that we see it is that AI is there to support people. It’s not to replace people.”
“So even this near full model that we’re working on with Home Guardian, it is to support the health professionals to better care for patients or for the elderly.
“It’s not to replace the health professionals. We will not replace and cannot replace the health professionals. We still need these people out there and these tools are just to support them.”