Toward Data-Driven Assessment of Caregiver’s Burden for Persons with Dementia using Machine Learning Models

Abstract

Giving care to persons with dementia (PwD) has a significant strain on the quality of life for familial caregivers. Due to the overdependent nature of PwD, caregivers are burdened with health issues, stress, depression, loneliness, and social isolation. As a result, there is a need for understanding the nature and severity of this burden. In this paper, we introduce a novel data-driven approach based on machine learning modeling to ascertain caregiver burden using multimodal data from multitudinal sources. In particular, we propose to leverage data from smart devices, wearables, and psychometric surveys, to assess caregiver burden employing both shallow and deep neural network architectures.