Detection of Social Isolation Based on Meal-Taking Activity and Mobility of Elderly People Living Alone

Objectives Background

Social isolation is probably one of the most affected health outcomes in the elderly people, particularly those living alone, due to the COVID-19 pandemic. Therefore, we try to identify it by detecting changes in the elderly such as malnutrition and lack of mobility.

Material and methods

The system consists of two types of sensors installed at various locations in the user's home: Passive infrared (PIR) sensors and reed switch sensors. It was implemented for 15 days in the home of a 26-year-old student living alone, as a first step to later be deployed in the home of elderly people.

Results

Our study showed strong similarities between the activities detected by the algorithm and the real activity pattern of the interviewed individual. In addition, the system was able to identify two daily patterns (weekday and weekend) of the person as he is a student and is present in class during the week.

Conclusion

A system composed of low-cost, unobtrusive, non-intrusive and miniaturized sensors is able to detect meal-taking activity and mobility. These results are an intermediate step in assessing the potential risk of social isolation in older people living alone based on these ADLs.