SCH: Robotic Caregiver to Comprehend, Assist, Relieve, and Evaluate for Patients with Alzheimer’s Disease (Robotic CARE for AD)

点击翻译按钮获取中文摘要

As the number and proportion of people aged 65 or older continue to increase in the United States, the number of Americans living with dementia is also growing. Among all dementia, Alzheimer’s disease (AD) and related dementia (ADRD) is the most common cause, accounting for approximately 60%-80% of the cases. Improving the care of people living with ADRD impacts the lives of not only the patients and their family but also the caregivers and their family, posing significant challenges to the well-being of the society. The purpose of this project is to exploit the autonomy and intelligence capability of a humanoid robot to comprehend, assist, relieve, and evaluate (CARE) patients with AD. The proposed Robotic CARE system has the capacity to detect the emotion and cognitive state of an AD patient, communicate and collaborate with the patient to accomplish basic instrumental activities of daily living (IADL) such as meal preparation, laundry, self-feeding, as well as help caregivers by reducing their levels of burden and stress. The proposed system could effectively bridge the CARE gap in our society, offering unprecedented potential for improving the quality of lives of both the patients and their caregivers. It is worth mentioning that the proposed Robotic CARE system is not to replace the caregivers, but rather to supplement the caregivers while providing companionship for AD patients since the system will learn to automatically assess patients’ cognitive state so as to better understand and respond to patients’ need. The CARE system is expected to advance the state-of-the-art in several related fields with its innovative and unique integration of cognitive assessment and rehabilitation with robot-assisted care. The proposed system makes contributions from three perspectives. First, cognitive assessment using instrumental activities of daily living (IADL) for patient-caregiver interaction will be conducted. Second, neurophysiological assessment using multimodal integration and unsupervised learning for patient-robot interaction will be conducted. Third, co-robot assessment using reinforcement learning for patient-robot-caregiver interaction will be conducted. The proposed team includes expertise from nursing with specialty in dementia care and gerontology, biomedical engineering with specialty in brain-computer interface, and computer science with specialty in machine/deep learning.