Smart Homes predicting the Multi-Domain Symptoms of Alzheimer ’ s Disease

As members of an increasingly aging society, one of our major priorities is the development of tools to early detect age-related disorders such as Alzheimer’s Disease (AD). The goal of this paper is to evaluate the possibility of using unobtrusively collected activity-aware smart home behavioral data to detect the multimodal symptoms of AD. After gathering longitudinal smart home data of 29 older adults for an average of > 2 years, we automatically labeled the data with corresponding activity classes and extracted time-series statistics containing 10 behavioral features. AD symptoms were assessed every six months by means of self-reported mobility, memory/cognition and mood tests. Using these data, we created regression models to predict AD symptoms as measured by the tests and a feature selection analysis was performed. Classification models were built to detect reliable absolute change in the scores predicting AD and SmoteBOOST algorithm was used to overcome class imbalance where needed. Results show that all mobility, cognition/memory and depression symptoms are predictable by the activity-aware smart home data, as well as a reliable change in mobility and visuospatial skills related to cognition. Results also suggest that not all behavioral features contribute equally to the prediction of every symptom. Future work must focus on improving the sensitivity of the presented models by collecting more longitudinal data and by focusing on class-imbalance suitable algorithms and in in-depth feature selection. The results presented herein contribute significantly towards the development of an early AD detection system based on smart home technology.