Professor Hwee Pink Tan
Associate Professor of Information Systems (Practice), Academic Director, SMU-TCS iCity Lab, Principal Investigator for the SHINESeniors project, Singapore Management University (SMU) iCity Lab
AI and Augmented Humanity
Since 2011, Tata Consultancy Services (TCS) and Singapore Management University (SMU) have been collaborating through the TCS-SMU iCity Lab to develop solutions for intelligent cities, including a pilot project deployed in the island nation of Singapore called SHINESeniors. The Government of Singapore provided the iCity Lab with a grant to enable researchers to work on technologies that non-intrusively monitor the well-being of elderly citizens who want to age in place in their own homes. The TCS-SMU iCity lab partnership extends until 2020 to focus on new research initiatives. Teams will work on building and piloting research-based solutions for smart cities considering citizen needs, experience, and response protocols.
The urban population today requires far more than basic health, education, and security amenities, and this is particularly true among the growing numbers of older adults who want to age in place in their own homes. Older adults require more care and more personalized services; the services they do receive are typically provided by a complicated and fragmented support system. It’s not uncommon for an older, independent adult to rely on a constantly changing mix of family members, home health care providers, local hospitals, and others. Although these service providers are helpful and often indispensable, they are also expensive and often inefficient. This fragmented, piecemeal approach results in too many manual interventions; many caregivers lack an overall picture of the needs and realities because they are focused on one particular need. The challenge also mandated non-intrusive monitoring.
Researchers at the iCity Lab piloted an Assisted Living solution, which combines sensor-enabled homes, personalised home care, and a medicine adherence care model. By using community assistance through a caregiver network and not the healthcare system, the Assisted Living solution helps control costs significantly, while still enabling the last-mile human touch.
Key features and functionalities:
- Unobtrusive sensors closely map daily activities; system events are generated based on periods of movement, rest, and ingestion of medication.
- In case of divergent data patterns, timely alerts tip-off community caregivers, who can step in to provide support.
- This discreet mode of tracking enables the elderly to live comfortably and independently while retaining a safety net – without feeling they are being monitored.
Monitoring is managed through dashboards, a portal for administrators, and a mobile app to alert caregivers if a situation requires intervention.
Based on the data captured by the sensors, a distinctive pattern emerges for each person. A narrative of their daily lives is created, which helps the team understand what’s normal and typical and deliver the right care at the right time. Community caregivers use a specially designed app interface to access the data and upload updates, and can also collaborate with other caregivers through the app. Also, the system puts no pressure on the residents – they don’t need to feed in any information manually. They simply live their lives.
The research team uses machine learning algorithms to predict cases of mild cognitive impairment, nocturia, and social isolation in the elderly, which can trigger the need for early assistance or intervention.
This was a pilot study, over three years, with a substantial population covering 90 homes with aging citizens living alone. Here are some representative results over that period:
- The system generated 17 yellow flag alerts when a resident/subject was away from home for more than 24 hours without informing caregivers.
- The system generated 19 inactivity events; two of these were people who were truly unwell and taken to a hospital.
- The system generated 363 emergency button alerts (though many of these were found to be requests for chat).
- The system generated medication box monitoring for 24 residents resulting in two personalized interventions.
The Assisted Living solution includes a Smart Data Hub to store the data, and custom-built Internet of Things (IoT) middleware, and an algorithm layer to dynamically analyze the data with a variety of machine learning approaches.
One of the key givens was the documented preference that older people prefer a consistent schedule and do not like to be disturbed or explicitly monitored. This compelled the iCity lab to plan carefully for the installation and maintenance of unobtrusive sensors that ensured residents comfort.
The number of older citizens is growing all over the world, and they represent a significant segment for businesses. Many solutions in serving this market are medical monitoring solutions, but ours is probably the first of its kind to focus on “Wellness and Living” through unobtrusive sensing and a community caregiver ecosystem. Our solution can easily and seamlessly support multiple types of sensors such as wearables if needed or desired. The 90 homes is a large installation. Using a Smart data hub, and learning on the data generated, we have been able to keep the residents satisfied with the solution.
The video of the project which is available on youtube captures the feelings of the subjects - https://www.youtube.com/watch?v=bXugFwHO7tI.
Interesting patterns and insights emerged from the project. For example, older people do not always sleep in the bedroom, but a few use the sofa. Also, the emergency button is not always used as an emergency button, but as an opportunity to chat with caregivers!
About Your Organization
Singapore Management University (SMU) collaborates with the TCS-SMU iCity Lab to develop solutions for intelligent cities such as a pilot project deployed called SHINESeniors. The Government of Singapore awarded the iCity Lab a grant to enable researchers to work on technologies that non-intrusively monitor the well-being of elderly citizens who want to age in place in their own homes.