Researchers are developing new autonomous technologies that help people make sure old remote family members are okay.
We live in a world where grandparents are increasingly distant from their relatives, with many moving for job opportunities.
There are now over 4 million people aged 65 and over living alone in the UK.
The series of national seizures during a pandemic coronavirus infection has made it even more difficult for people to ensure that these old loved ones are safe and healthy.
Families want peace of mind that their old relatives are living well, and maintaining a relatively active lifestyle. While we can call on them to look at them, that is often not enough.
Dr Rebecca Killick, Principal Investigator of the study from Lancaster University, said: “We can make the call using telephones, or video conferencing devices, but we often find that many older people says they are well, and we are not sure. You will try to do more research but it is difficult when you do not have the social ads that confirm your suspicions. “
To help overcome these problems, the researchers, from Lancaster University, are working with technology company Howz to devise a new way to interpret statistical data from smart home sensors.
These sensors, which act very similar to home security sensors, record movement and other tasks such as opening and closing doors and using appliances.
The researchers ’modern statistical analysis methods involve representing data visually in a circular graph that can be placed in sequences, rather than in a linear format. Traditional statistical methods struggle with nature from time to time in people’s daily habits – with active times, when making food or drinks, and less active times, when sleeping, reading or watching TV.
The new method transcends previous boundaries in identifying changes, using what statisticians refer to as ‘changepoint analysis’.
The sensor system from Howz, along with the insights gained from the researchers ’new statistical approach, was used in the homes of four study participants for 56 days.
The results show that the system was able to identify broad patterns in a day, such as hours in the day when participants were awake and asleep, as well as times when they were performing activities. as food preparation.
In addition, the new statistical method made it possible for them to more easily detect and compare participants’ practices over several days. This is important because it lays the foundation for systems that quickly recognize small changes in behavior that may indicate a decline in health.
This search allows us to identify daily, or weekly, carrier patterns and unforeseen trends. This can be used as a barrier for family members, or carers, to have a conversation. “
Dr Rebecca Killick, Senior Research Fellow, University of Lancaster
“Our innovative approach to the problem of identifying patterns of behavior is to make the data worse day by day, and over time more subtle changes are found. This is crucial because we can -now detect earlier changes in behavior or capture changes that would have been missed in the past. “
The researchers believe that their new statistical approach to points of change could be useful in other areas of research, including the environment, official statistics and industry.
Source:
Magazine Reference:
Taylor, SAC, et al. (2021) Assessing daily patterns using home activity sensors and within a detection point of change. Royal Statistical Society: Series C.. doi.org/10.1111/rssc.12472.