Smart' Dresser Prototype Guides People With Dementia in Getting Dressed

A study published in JMIR Medical Informatics details how a smart home prototype could help people with dementia dress themselves via automated assistance, reports ScienceDaily . The DRESS system combines sensors and image recognition to track progress using barcodes on apparel to identify the type, location, and orientation of a piece of clothing. DRESS involves structuring a five-drawer dresser topped with a tablet, camera, and motion sensor, with one piece of clothing per drawer, to follow an individual's dressing preferences. A skin conductance sensor worn as a bracelet tracks a subject's stress levels and related frustration. A caregiver initiates DRESS and monitors progress from an app, while the person with dementia receives an audio prompt in the caregiver's voice to open the top drawer, which lights up. The clothes in the drawers have barcodes that are detected by the camera, and if an item is put on properly, the system prompts the person to move to the next step. Should the system detect an error or lack of activity, audio prompts are used for correction and encouragement. If it detects ongoing issues or an increase in stress levels, the system can alert a caregiver to provide assistance. The prototype can identify clothing orientation and position, and deduce one's current state of dressing. In initial phases of putting on shirts or pants, DRESS accurately detected participants' clothing 384 of 388 times. However, it could not consistently identify when one completed donning an item of clothing, missing these final cues in 10 of 22 cases for shirts and five of 22 cases for pants.