June 6, 2017 | Written by: Susann Keohane
Categorized: AI/Watson | IBM: The Backstory
Share this post:
My grandmother suffered from an aneurysm when she was in her 80s. Not only did she survive, but she learned to walk and talk again, living well into her 90s. As a little girl, I remember thinking she was so incredible for having such determination to live the best possible life given those daunting challenges.
Today, my grandmother inspires my work. As the IBM Global Research Leader for the Strategic Initiative on Aging, I explore how cognitive technology and the Internet of Things (IoT) can be combined to improve quality of life for the elderly.
By 2020, elderly people will outnumber children for the first time in history. With this demographic boom, it’s the right time and the right problem to solve.
Staying at home longer
As the population ages, the nature of care is being transformed—moving beyond medical staff, hospitals and nursing homes, to also include family and community assistance while living healthier and longer at home.
Ambient assisted living, which leverages the data created by sensors and devices throughout a home or facility to improve quality of life, can facilitate this transformation. By 2022, each typical family home will contain an average of 500 smart devices. Caregivers can use streams of data from these devices to identify patterns that tell the story of daily life. For instance, when someone gets out of bed, when they cook and when they’re moving about, and how often they stay connected with their family, friends and community.
Each of us will have our own pattern, and learning these patterns is key for aging in place safely and mitigating risk. Small variations over time may signify changing needs, while sharp, abrupt fluctuations can alert to a problem.
Alerts, generated through passive monitoring of IoT devices, can notify caregivers and family members of these small or abrupt changes. For example, if John’s aging mother hasn’t made a cup of tea by 10 a.m., that’s quite unusual. So, John receives an SMS message to tell him that his mother might not have gotten out of bed yet. Even if she’s fine, she’d love to hear from her son checking that she’s okay.
Each of us will age uniquely based on genetics, environment and personal attitude towards life. As we age, we acquire physical and cognitive disabilities. Being able to proactively monitor changes in our abilities over time is central to successful care.
Combining cognitive and IoT technologies enables us to glean deeper insights into these changes by integrating and understanding sensor data and additional sources such as video, audio, biometric sensors and soon possibly even smells. Data from electronic medical records (EMR) can also be added to provide insight on context, thereby connecting what is happening in the environment to a person’s physical health.
In just a few months, there may be over 500 million data points on just one individual. So, machine learning capabilities, like those enabled by cognitive technology, are critical to analyzing and modeling the data.
Together, these technologies can provide a very comprehensive picture of our overall well-being. They will help ensure that vital life decisions, such as the right time to stop driving or transition to a different level of care, are based on evidence from data and not on emotions or fear.
Improving assisted care
In nursing homes and assisted living facilities, medical staff and care providers can better understand how well a person is recuperating by correlating sensor and EMR data.
In the state of Oregon for instance, we’re working with the Avamere Family of Companies to gain insights into physical and environmental conditions. We’re also working to better understand the factors that affect 30-day hospital readmission rates in patients. We’re monitoring movement, air quality, gait analysis, fall risk factors, and daily activities, like personal hygiene, sleeping patterns and trips to the bathroom. By applying advanced analytics and machine learning techniques to sensor and health data, the goal is to create and maintain a contextualized understanding of residents and their health.
This type of contextualized pattern identification informs predictions, and is enabled by pulling silos of data together to make it meaningful and actionable. Instead of waiting for an event to happen, we have triggers that can alert us when there’s an increased probability for it to happen before it takes place. This provides opportunities for putting preventative measures in place—taking eldercare to the next level.
Whether at home or in a facility, we’re using cognitive technology and IoT to help everyone feel younger, age gracefully, and follow in my grandmother’s footsteps.
For more details, please read our “Outthink Aging” and “Loneliness in Aging” reports, and visit our website.