Home Automation and Smart Homes diffusion are providing an interesting opportunity to implement elderly monitoring. This is a new valid technological support to allow in-place aging of seniors by means of a detection system to notify potential anomalies. Monitoring has been implemented by means of Complex Event Processing on live streams of home automation data: this allows the analysis of the behavior of the house inhabitant through quantitative indicators. Different kinds of quantitative indicators for monitoring and behavior drift detection have been identified and implemented using the Esper complex event processing engine. The chosen solution permits us not only to exploit the queries when run “online”, but enables also “offline” (re-)execution for testing and a posteriori analysis. Indicators were developed on both real world data and on realistic simulations. Tests were made on a dataset of 180 days: the obtained results prove that it is possible to evidence behavior changes for an evaluation of a person’s condition.
The most frequent form of dementia is Alzheimer’s Disease (AD), a severe progressive neurological pathology in which the main cognitive functions of an individual are compromised. Recent studies have found that loneliness and living in isolation are likely to cause an acceleration in the cognitive decline associated with AD. Therefore, understanding social behaviours of AD patients is crucial to promote sociability, thus delaying cognitive decline, preserving independence, and providing a good quality of life. In this work, we analyze the localization data of AD patients living in assisted care homes to gather insights about the social dynamics among them. We use localization data collected by a system based on iBeacon technology comprising two components: a network of antennas scattered throughout the facility and a Bluetooth bracelet worn by the patients. We redefine the Relational Index to capture wandering and casual encounters, these being common phenomena among AD patients, and use the notions of Relational and Popularity Indexes to model, visualize and understand the social behaviour of AD patients. We leverage the data analyses to build predictive tools and applications to enhance social activities scheduling and sociability monitoring and promotion, with the ultimate aim of providing patients with a better quality of life. Predictions and visualizations act as a support for caregivers in activity planning to maximize treatment effects and, hence, slow down the progression of Alzheimer’s disease. We present the Community Behaviour Prediction Table (CBPT), a tool to visualize the estimated values of sociability among patients and popularity of places within a facility. Finally, we show the potential of the system by analyzing the Coronavirus Disease 2019 (COVID-19) lockdown time-frame between February and June 2020 in a specific facility. Through the use of the indexes, we evaluate the effects of the pandemic on the behaviour of the residents, observing no particular impact on sociability even though social distancing was put in place.
Wellness assessment refers to the evaluation of physical, mental, and social well-being. This work explores the possibility of applying technological tools to assist clinicians and professionals to improve the quality of life of people through continuous monitoring of their wellness. The contribution of this paper is manifold: a coarse-grained localization system is responsible for monitoring and collecting data related to patients, while a novel wellness assessment methodology is proposed to extract quantitative indicators related to the well-being of patients from the collected data. The proposed system has been installed at “Il Paese Ritrovato", an innovative health-care facility for Alzheimer’s in Monza, Italy; first satisfactory results have been obtained, and the dataset shows great potential for several applications.
The detection and monitoring of a sitting or lying person using non-intrusive sensing technologies can contribute to a wide range of applications aimed at improving quality of life, such as human activity recognition, sleep quality assessment, and the detection of behavioural abnormalities. This systematic review provides a comprehensive overview of non-intrusive sensing technologies for detecting sitting and lying individuals, including those capable of monitoring vital signs. The article offers a detailed classification of these technologies based on the physical variables they measure, providing a concise understanding of the principles behind each technology, along with an analysis of their respective pros and cons. For each technology, this review provides an overview of the most significant recent works (of 552 analyzed, 90 were selected) that demonstrate their applications and purposes. The spectrum of sensors analyzed includes mechanical, electrical, optical, acoustic, magnetic, and thermal sensors, each evaluated for its strengths, limitations, and potential uses, showing recent trends.
Population aging requires innovative solutions to increase the quality of life and preserve autonomous and independent living at home. A need of particular significance is the identification of behavioral drifts. A relevant behavioral drift concerns sociality: older people tend to isolate themselves. There is therefore the need to find methodologies to identify if, when, and how long the person is in the company of other people (possibly, also considering the number). The challenge is to address this task in poorly sensorized apartments, with non-intrusive sensors that are typically wireless and can only provide local and simple information. The proposed method addresses technological issues, such as PIR (Passive InfraRed) blind times, topological issues, such as sensor interference due to the inability to separate detection areas, and algorithmic issues. The house is modeled as a graph to constrain transitions between adjacent rooms. Each room is associated with a set of values, for each identified person. These values decay over time and represent the probability that each person is still in the room. Because the used sensors cannot determine the number of people, the approach is based on a multi-branch inference that, over time, differentiates the movements in the apartment and estimates the number of people. The proposed algorithm has been validated with real data obtaining an accuracy of 86.8%.