Smart sensors used to enhance home care for patients
October 2, 2017
Caring for seniors at home is about to get a whole lot smarter. With recent census data indicating one-person households surpass all other types of living situations in Canada, in part due to longer life expectancy, the need to help aging Canadians maintain independence and quality of life is stronger than ever.
The result, says Alex Mihailidis, scientific director at AGE-WELL NCE Inc., is a “grass roots” movement to adopt – and adapt – smart technology.
“It started to emerge over the last few years and is really starting to take hold now,” said Mihailidis, noting that consumers are gravitating to off-the-shelf control devices such as smart thermostats, lighting, security systems, watches, cameras and wearable fitness trackers as a means to keep seniors healthy and safe at home.
“It’s not the companies themselves that are pushing to be in this market, it’s caregivers, families and seniors themselves making use of these tools for their own purposes,” he said.
The driving force behind adoption is the ongoing evolution and maturity of the Internet of Things (IoT), the interconnection of embedded computing devices (sensors) in everyday objects so that they can send and receive information. For researchers at AGE-WELL, a federally funded Networks of Centres of Excellence (NCE) program hosted by the University Health Network in Toronto, the opportunity stretches far beyond consumer products.
Launched in 2015, the pan-Canadian network brings together researchers, non-profits, industry, government, care providers, older adults and caregivers to find innovative ways to support independent living and enhance wellness among seniors. Many of their projects are examining how digital sensors can be used to collect and monitor health data, with a focus on finding non-invasive ways to install sensors in a home.
One example is a joint research effort led by Dr. Frank Knoefel, physician and clinical scientist at Ottawa’s Bruyère Research Institute and Rafik Goubran, professor and dean, Faculty of Engineering and Design, at Carleton University in Ottawa. As a specialist in geriatric rehabilitation, Dr. Knoefel was initially interested in applying smart technology to function as a type of early warning system for senior patients following hip surgery.
The goal was to embed a sensor between a patient’s mattress and box spring so clinicians could remotely monitor patient movements getting in and out of bed.
“If we could do that, it would give us a level of confidence that if they got out of bed alright, there’s a good chance they’d make it through the rest of the day alright; whereas if that first transfer in the morning was already wobbly, then we’d be more concerned,” explained Dr. Knoefel.
The resulting ‘smart mat’ is a pressure sensitive foam pad that uses fibre optic light to measure pressure. As patients manoeuver in and out of bed, pressure sensors collect information related to their movements, enabling researchers to use that data to remotely monitor things like how long it takes them and how many times they stand up and then sit back down again. The technology has since been refined to also measure breathing rates and most recently, to measure fluid retention, an early indicator of trouble in patients with chronic heart failure.
“It’s really incredible that the technology can do this,” said Dr. Knoefel. “It makes us think completely differently about what we could be able to do in peoples’ homes.”
At first, the research team focused on modifying an existing pressure sensitive mat, commonly used in car seats to engage seatbelt indicators. Moving forward, they are developing a custom solution and are currently working to obtain a patent for their technology. Dr. Knoefel says it’s not a question of will the technology be commercially viable, but how and when.
“It becomes a question of who do we partner with,” he said, adding that one possibility would be home-care providers who could use the mat as a form of smart triage. “If they had a case load of 40 frail people, it would be useful if first thing in the morning they could turn on their computer and see three people who didn’t get up at their usual time or got up really wobbly, or here’s someone who just gained a kilogram over two days … they’re about to go into heart failure,” he explained.
Saint Elizabeth, a national social enterprise providing home care, health solutions and education is investing in remote sensing research as it moves forward with its IntelligentCare framework, a care delivery model designed for a digital age.
All patient encounters, including virtual visits, direct visits and remote patient monitoring, will be tracked and maintained in a single longitudinal health record.
The framework is divided into four components: Self Care, an online service for patients looking to do self-guided symptom assessment or to access community resources; Tele Care, a clinical patient support centre that provides care navigation and clinical triage services on demand; Tele Health, a service that leverages remote patient monitoring devices and two-way communication channels to monitor patients at home; and Smart Home, a service that will provide remote sensing and other smart home health technologies to enable seniors and medically fragile patients to live independently.
Over the next 12 years, Saint Elizabeth will invest $12 million in research with much of it focused on remote sensing, says Barry Billings, IntelligentCare Architect, Saint Elizabeth. “We’re thinking of it as a patient support program,” he said, noting that the focus is on healthcare first, technology second.
“Somebody might say that (remote) technology should fall in the telehealth bucket. Yes, it’s telehealth software, but what we deliver is far more than telehealth. It’s telehealth with care pathways that are co-developed with our partners.”
The goal of the Smart Home program, still under development, is to use motion sensors and other “unobtrusive, passive sensing” technologies, he says, to monitor the activity and normal behaviour patterns of those living independently. The data collected is sent to a smart cloud and predictive analytics are applied to identify variations in behaviour so that clinicians, or even other smart technologies, can intervene.
“If someone gets up in the middle of the night and we know from the last 48 days that their next step is the kitchen to get a glass of water, why don’t we turn on a light in the kitchen?” suggested Billings. “If we can prevent a fall one out of 1,000 times, we’ve done a really good job.”
Another area of research is focused on developing a digital health assistant, similar to the way Amazon Alexa or Google Home function.
Using natural language technology, the device becomes the interface to connect patients, family members, caregivers and the wider healthcare team. In addition to being used for passive monitoring, it collects self-reported symptoms and can even be used to manage appointments.
“If Dad wants to get a haircut, he can ask it to put haircut on his calendar. The message goes out to two or three family members and somebody goes to their smartphone and says I can take Dad Friday morning,” he described.
Saint Elizabeth is also exploring how Tess, an artificial intelligence (AI) chatbot created by X2AI Inc., can be used to support patients and family caregivers at home. Tess works by holding conversations with patients and, with the oversight of a therapist, provides coping mechanisms and other self-care supports. One idea is to use the technology as a virtual emotional coach to calm people who are palliative and have questions about death and dying.
According to analysis from Frost & Sullivan, the real role of AI systems is to augment the expertise of trained clinicians. The capacity for AI technologies to extract information from various sources, including remote sensors, and turn it into actionable data is allowing researchers to tackle challenges that “previously had no other means of recourse,” the company stated.
“By 2025, AI systems could be involved in everything from population health management to digital avatars capable of answering specific patient queries,” noted Frost & Sullivan Transformational Health Industry Analyst Harpreet Singh Buttar. “On a global scale … AI is expected to play a significant role in democratization of information and mitigating resource burdens.”
An AGE-WELL research team at the University of Western Ontario in London is applying advanced telerobotics to design a system that will make it easier to deliver individualized therapy at home for older stroke survivors and seniors with age-related movement disorders.
Co-led by Rajni Patel, professor and Canada research chair in advanced robots and control, and Dr. Mandar Jog, a movement disorders neurologist, the team is aiming to bring stroke therapy to a wider catchment of patients. Their remote rehabilitation platform combines haptic or force-enabled robotic technology, virtual reality and Internet communication and is similar to a virtual gaming experience.
In one of the exercises, patients use a haptic robotic arm to move a ball on a computer screen, according to prompts. As they make the movements, they feel resistance from the ball; a therapist – who can be sitting next to them or in an office far away – uses a similar robotic arm to control a second ball on the screen. The therapist can create a sense of lag so that the patient feels more resistance and hence has to work harder.
“In a situation where a person has gone through therapy in the clinic and goes home, we’d like to continue the therapy as much as possible,” said Patel. “Ultimately what we’re trying to do with post-stroke rehabilitation is rebuild some of the neuro connections in the brain … the more actions you do, the stronger the connections get.”
The innovative system is ready for patient evaluation and the team expects to have a commercial partner by the end of the year.
In addition to passive remote sensors and AI technologies, wearable devices are also emerging as a viable means to monitor seniors at home. However, rather than using prompts and alerts to engage users, similar to the Fitbit approach, researchers are targeting more passive applications.
An AGE-WELL team at University of Toronto (UofT), for example, is developing WearCOPD, a wearable application to monitor chronic obstructive pulmonary disease (COPD) patients at home. The goal is to monitor their condition and detect early exacerbations – such as shortness of breath, worsening cough or a decrease in activity level – so that a timely intervention can occur before a trip to the emergency room is necessary.
Right now the team is using a standard Android smartwatch to monitor heart rate, activity level and collect audio (to detect coughing). Approximately 30 COPD patients from Toronto General and Sunnybrook Hospitals wore the watch for three months, generating well over 1,000 hours of patient data which is being analyzed. Coughing analysis is complete and now the team is working to ensure the device provides an accurate measure of heart rate.
“One of the next steps is validating how good the consumer smartwatches are,” said project lead Dr. Robert Wu, associate professor in the Faculty of Medicine at U of T. “The other part is asking participants what they would like to see and what they would use … it’s about providing the appropriate technology to patients in the form they want, that can actually help them.”
At Saint Elizabeth, Billings refers to growing smart technology trend as “augmented intelligence.” Machines and devices may be capable of collecting and processing more data in more ways than ever, but human intervention is still required, he says.
“We believe in having all of that sensor data processed and sent to somebody who’s clinically trained to say 97 times out of 100 that is the right diagnosis, treatment or intervention,” said Billings.
“But the one time that it’s not, or the time the sensors malfunction, or there’s missing data, it needs to function as a healthcare assistant for the healthcare worker.”