Predictive model for elderly dependency assessment in ambient assisted living

摘要

A significant proportion of elderly population suffers from age-related health issues and different chronic diseases which leads to a progressive decline in physical and cognitive skills which prevents them to live independently in their home and to perform basic Activities of Daily Living (ADL). In this context, the integration of information and communication technologies (ICT) has resulted in a rapid emergence of pervasive healthcare systems such as Health Smart Homes which aim to monitor and evaluate the person’s health condition and their behavior in performing ADL and to avoid, as long as possible, the delays of recourse to healthcare institutions (e.g. nursing homes and hospitals). In this research, we propose a predictive model for detecting behavioural and health-related changes in a patient who is monitored continuously in an assisted living environment. We focus on keeping track of the evolution of the dependency level and detecting the loss of autonomy for elderly person using a Hidden Markov Model based approach. We were interested in including blood pressure in this predictive process as it is considered the primary risk factor for cardiovascular disease, stroke, kidney failure and many other diseases. Our simulation was applied on a empirical dataset that concerned 3046 old persons monitored over 9 years. The results show that our model accurately evaluates person’s dependency, follows its evolution of autonomy over the time and thus predicts moments of important changes.