RadioMe: detecting agitation using smartwatches
Research type
Research Study
Full title
RadioMe: Understanding biometric data as a warning sign of pending agitation in dementia and in response to preferred music listening
IRAS ID
327853
Contact name
Alexander Street
Contact email
Sponsor organisation
Anglia Ruskin University
Duration of Study in the UK
1 years, 0 months, 2 days
Research summary
People living with dementia independently in their own home can experience stress and agitation. Agitation is one of the major symptoms that leads to care home admission and some research has shown that listening to preferred music can help manage it. The larger study that this proposed one would be part of aims at designing a system that will automatically play preferred music at times when it may bring benefit - when the user experiences stress as indicated by changes in their biometric data (e.g. heart rate or HR.
In order to build a machine learning model for the RadioMe system that will enable it to calibrate accurately and play preferred music playlists when each user's data indicates agitation, we need to gather biometric data from people with dementia when they experience it and feed it into a machine learning model. This HR data acts as the 'on/off' switch for the preferred music. We are collecting these data with current RM participants, but we want to maximise the generalisability of the RM system by including those with more frequent and severe agitation.
Participants will be recruited from NHS trusts and care homes within England. The settings will be chosen on the basis of having groups of people with dementia with agitation (e.g., wards). Members of the healthcare team will facilitate initial contact with potential participants. Researchers will then recruit those interested into the study.
Twenty people on the wards will participate for six-weeks maximum, with up to 12, 60-minute periods of wearing the smartwatch on their wrist and being observed in social areas on the sites by researchers. There will be up to six, 30-45-minute music listening sessions in a designated room (also wearing the monitor). Biometric data will be anonymous and stored on the password protected RadioMe system.REC name
London - South East Research Ethics Committee
REC reference
23/LO/0593
Date of REC Opinion
7 Aug 2023
REC opinion
Further Information Favourable Opinion