Alzheimer’s Garden: Understanding Social Behaviors of Patients with Dementia to Improve Their Quality of Life

Abstract

This paper aims at understanding the social behavior of people with dementia through the use of technology, specifically by analyzing localization data of patients of an Alzheimer’s assisted care home in Italy. The analysis will allow to promote social relations by enhancing the facility’s spaces and activities, with the ultimate objective of improving residents’ quality of life. To assess social wellness and evaluate the effectiveness of the village areas and activities, this work introduces measures of sociability for both residents and places. Our data analysis is based on classical statistical methods and innovative machine learning techniques. First, we analyze the correlation between relational indicators and factors such as the outdoor temperature and the patients’ movements inside the facility. Then, we use statistical and accessibility analyses to determine the spaces residents appreciate the most and those in need of enhancements. We observe that patients’ sociability is strongly related to the considered factors. From our analysis, outdoor areas result less frequented and need spatial redesign to promote accessibility and attendance among patients. The data awareness obtained from our analysis will also be of great help to caregivers, doctors, and psychologists to enhance assisted care home social activities, adjust patient-specific treatments, and deepen the comprehension of the disease.

Understanding Social Behaviour in a Health-Care Facility from Localization Data: A Case Study

“…As in [ 27 ], we extend the definition of the to capture the sociability of an individual dweller. Considering a user as active on day d when we can track their position throughout day d , we define the Individual as: …”

“…Social interactions are often bound to the places where they occur. In our previous work [ 27 ] we introduced a measure of the popularity of places, called the Popularity Index. Considering the context of Alzheimer’s Care Home Facilities, our definition of “popular place” can be given in light of this statement—the more a place is frequented, the more it can be considered popular.…”

“…Through an in-depth analysis of the correlation between the social behaviour of individuals measured through the Individual and other variables, in [ 27 ] we observed that patients’ social behaviour depends on the season, the temperature, the dweller’s bedroom floor, the predisposition to walk on a certain day and the patient’s average hourly walked distance. Given some of these variables, we aim to exploit the observed dependence and predict the social behaviour of the whole community.…”

“…As in [27], we extend the definition of the RI to capture the sociability of an individual dweller. Considering a user as active on day d when we can track their position throughout day d, we define the Individual RI as: Since the RI for an individual over the entire day is computed as the mean of squares, spending a certain amount of time with a single person is valued more than spending the same amount of time cumulatively with more people.…”

“…Our purpose is to analyze these localization data, build visualization tools to observe and comprehend the indexes of sociability and popularity introduced in previous works [ 13 , 27 ] and design predictive models for their estimation over time.…”