The Irish robot helping older people live independently

Gerontechnology will play a crucial role in care as fewer people opt for careers in sector

The GaitKeeper app  records 60-frames-per-second video footage of someone as they perform a walking test, capturing 25 core joint locations on the body. Within minutes the AI engine calculates elements of gait that turn out to be important predictors of both physical and cognitive decline.

The GaitKeeper app records 60-frames-per-second video footage of someone as they perform a walking test, capturing 25 core joint locations on the body. Within minutes the AI engine calculates elements of gait that turn out to be important predictors of both physical and cognitive decline.

 

If you haven’t yet heard of gerontechnology or the “silver tech” space then you soon will. It will become a core aspect of both younger Boomers and Gen X’s care as they age. Silver tech encompasses everything from devices that aid independent living to apps that can predict dementia, and robots to keep us company.

This market will be worth $915 billion (€777 billion) by 2025, according to analyst firm BBC Research. It’s no wonder: there were 703 million persons aged 65 years or over in the world in 2019 and this is projected to double to 1.5 billion by 2050.

In Ireland, there is a newly-formed Gerontechnology Hub at Tallaght University Hospital (TUH) led by consultant geriatrician Prof Sean Kennelly. He says it is vital that with longer lifespan comes a corresponding increase in health span. And technology can support this.

“One of the core pillars of TUH is technology and innovation. We have a very supportive executive group who have encouraged collaborations with tech companies to put together an expert group of developers, innovators, clinicians and patients,” explains Kennelly.

One of the first collaborations has been with Dublin City University (DCU) spin-in GaitKeeper. This app records 60-frames-per-second video footage of someone as they perform a walking test, capturing 25 core joint locations on the body. Within minutes the data is processed by the AI engine and used to calculate elements of gait – such as stride length or heel strike – that turn out to be important predictors of both physical and cognitive decline.

GaitKeeper received funding from Enterprise Ireland in 2019 to spin into DCU in order to develop a robust healthcare application for their tech. It has been part of a clinical study at TUH since November 2020, where 250 gait tests comprising 250,000 images have been processed through the AI engine, says chief executive Aidan Boran.

“The TUG (timed up and go) test is most frequently used to understand fall risk in older people. The patient is asked to move from a seated position, walk three metres and return all the while measuring the length of time it takes,” explains Boran.

“Tens of thousands of these tests are performed in Ireland on a yearly basis but it is mainly a manual process using a stop watch or similar while making notes, which has an element of subjectivity.”

There is existing tech out there to measure gait. The Vicon motion capture system – familiar to anyone who watched Andy Serkis wear a CGI suit to play Gollum in The Lord of the Rings – can triangulate the person down to the millimetre. But with high accuracy comes high cost so this isn’t always ideal in a clinical setting.

Affordable

Other systems might be more affordable but still require room for specialist equipment and an IT expert on hand to set it up. GaitKeeper aims for an “anytime, anywhere, anyone” approach with a smartphone app that allows the clinician to get it done in a few taps. It has been validated against the more expensive gait systems to ensure comparable accuracy.

Business Today

Get the latest business news and commentarySIGN UP HERE

“Being a start-up in this area is different to the traditional route. The concept of MVP (minimal viable product) is not applicable. Who wants an MVP stent in their heart? It has been a long road of developing this product but we are now a regulated Class 1 medical device,” notes Boran.

What these kinds of technologies excel at is putting into concrete data what clinicians have long observed but only intuited: “Based on the walk from a waiting room to the door of the consult room, many clinicians will have identified the pertinent issues before someone even sits down to talk with them,” explains Kennelly.

“How we walk is a very good indicator of our cognitive and psychological performance – just think about how we walk when we’re feeling confident or feeling sad – but we’ve never had a means of easily measuring this or extracting the data before.”

Kennelly is also working with a tech company specialising in analysing human speech patterns because, just like walking, speech is incredibly complex and a slight change can be indicative of a physical condition like Parkinson’s, a cognitive disease like dementia and even the psychological – depression or anxiety.

“Myself and colleagues in Trinity College Dublin (TCD), including Prof Richard Reilly who is in the area of neurological engineering, are particularly interested in this in relation to dementia detection. We will compare the AI’s speech analysis to that of clinical professionals like speech therapists as well as a manual analysis of digital recordings.”

Kennelly is similarly interested in the application of robotics for independent living, name checking Stevie II, the TCD-developed healthcare robot that made the cover of Time magazine in 2019 and monitoring companion assistant Mylo, created by Candace Lafleur.

When she was only 32 years old Lafleur had a massive stroke. The Canadian businesswoman stayed in hospital for several weeks because, although she was recovering well and didn’t need much medical intervention, 24/7 supervision was required before she could return to her young family.

“I noticed there were other women on my ward who were older than me. At home, if they fell, they might not be able to reach their medical alert bracelet or other alert system. It was horrifying that they had to stay in care permanently just because there was no one to monitor them in their house. So Mylo came out of that,” reflects Lafleur.

“So I just started doodling an idea for a robot assistant from my hospital bed. It’s like that Lily Tomlin quote, ‘Someone has to do something about this’. And then you realise that you are someone.”

Live independently

Mylo is a support and companion robotic assistant whose primary role is to monitor older people so that they can live independently. He does a little bit of everything from reminding his owner to take their medication (complete with an image of where it’s kept) to initiating video calls with family and friends. His rental cost comes with in-home set-up to start, and remote set-up for the key contact person who learns how to operate Mylo from their laptop or phone.

Most importantly, Mylo allows family members to non-intrusively monitor their parent or grandparent. This robot allows for installation of third-party apps including fall detection. In the event of a fall, their contact is notified and can remotely drop into Mylo to see what’s happening.

“Say, for example, your mother is living by herself and you want to check in on her. She has early stages of dementia and is having issues with independence. You can avatar into Milo from your phone and as he drives around her house you can see what he sees and hear what he hears,” explains Lafleur.

“You can check if your mum locked the door or ensure home care has arrived without having to be at her house constantly. And she can have that independence as well.”

Mylo has been trialled in a nursing home in Louth where Lafleur and her team are seeing how these robots are used on a daily basis. He was even brought into a nearby primary school to allow local pupils to get to grips with robotics and coding.

“The nursing home trial was great until suddenly the data went dark. I thought ‘Oh god, do they hate Mylo? Is he locked in a closet somewhere?’ It turns out one of the pupils had a grandparent there and taught the nurses how to back door out of his system to access YouTube. Mylo wasn’t making video calls or moving about much because the residents were binge-watching Daniel O’Donnell videos.”

Despite the wee Daniel hiccup, Lafleur says the robot is most popular for voice-activated video calls because many older people in residential care don’t have phones or certainly not smartphones. And when Mylo is done facilitating a call he goes off to charge himself before returning for his next command.

With fewer people opting for careers in care there is a need to supplement this role for the world’s aging population. The healthcare-cum-companion robot will be more common in the near future, says Lafleur. The only difference is we’ll asking Mylo to play some Oasis or Blur.

The Irish Times Logo
Commenting on The Irish Times has changed. To comment you must now be an Irish Times subscriber.
SUBSCRIBE
GO BACK
Error Image
The account details entered are not currently associated with an Irish Times subscription. Please subscribe to sign in to comment.
Comment Sign In

Forgot password?
The Irish Times Logo
Thank you
You should receive instructions for resetting your password. When you have reset your password, you can Sign In.
The Irish Times Logo
Please choose a screen name. This name will appear beside any comments you post. Your screen name should follow the standards set out in our community standards.
Screen Name Selection

Hello

Please choose a screen name. This name will appear beside any comments you post. Your screen name should follow the standards set out in our community standards.

The Irish Times Logo
Commenting on The Irish Times has changed. To comment you must now be an Irish Times subscriber.
SUBSCRIBE
Forgot Password
Please enter your email address so we can send you a link to reset your password.

Sign In

Your Comments
We reserve the right to remove any content at any time from this Community, including without limitation if it violates the Community Standards. We ask that you report content that you in good faith believe violates the above rules by clicking the Flag link next to the offending comment or by filling out this form. New comments are only accepted for 3 days from the date of publication.