Dialogue-Based COACH Project (Intelligent Supportive Environment for Older Adults)
Keywords: Cognitive device, cognitive orthosis, smart homes, assisted cognition, context-aware design, ADL prompting, ADL guidance, dialogue-based system, sensorless monitoring
Overview of Research
Adults with Alzheimer’s disease (AD) and other types of dementia often have difficulty performing basic activities of daily living (ADL) such as bathing, feeding and toileting. Consequently, they become highly dependent on formal or informal caregivers. As caregivers assume responsibility for a person with progressive dementia, they may experience increased stress and burden. Moreover, dementia patients can feel frustrated due to loss of autonomy.
The COACH (Cognitive Orthosis for Assisting with aCtivities in the Home) is a prototype for an intelligent supportive environment which was developed to assist people with dementia complete ADLs with less dependence on a caregiver, representing one of the first clinically tested supportive devices to use artificial intelligence techniques. To date, prototypes of the COACH system have completed clinical trials based around the ADL of handwashing with subjects who had moderate-to-severe dementia. Currently, the system uses automated hand-tracking to monitor the user and provides recorded prompts when an error is detected. COACH provides up to four prompts with increasing levels of support. The types of prompts include a low-guidance verbal prompt, high-guidance verbal prompt, a prompt with video demonstration and a call to the caregiver. However, there are certain limitations associated with the use of sensors in the COACH system. For instance, installing sensors in a dementia patient’s home can be an expensive and inconvenient process. Sensors are also typically customized for specific tasks and cannot be generalized to others.
This study aims to investigate the development of a spoken dialogue-based prompting system for ADL assistance, specifically with the task of handwashing. The prototype will be extension of the COACH system but instead of the hand-tracking sensor, it will solely rely on spoken-dialogue to obtain input data for the underlying COACH model. The system will be created by integrating off-the-shelf natural language processing tools with the COACH model. DialogFlow by Google has been chosen for this project due to its completeness and its ability to support integration with platforms such as Google Home and Amazon Alexa. An experimental study will be performed to investigate whether spoken language input can be used by the existing COACH model in place of the hand-tracking sensor to estimate user state and environment during the progression of handwashing. The study will be conducted with young adults who will interact with the system while washing their hands. This study will help us improve the usability of the existing COACH system by introducing speech-based interaction. Moreover, it will shed light on whether spoken dialogue can replace the role of sensors in CATs, providing a more natural method of communication while helping dementia patients with ADL performance.
Figure 1. Spoken Dialogue System Architecture
Research Team
Neeraja Dharan, University of Toronto
Alex Mihailidis, University of Toronto
