[FTS4–01–01]: DPUK WEARABLES PLATFORM

Background: With recent technical and commercial advances in wearable computing we are now in a position where cheap, off-the-shelf devices, such as fitness trackers and smartwatches, offer instrumentation and performance comparable to expensive, specialised equipment from even a decade ago. Such devices can now be used for the collection and analysis of fine-grained personalised data and to scale studies utilising this data across large populations. The type of data captured, for example longitudinal movement data, sleep pattern analysis and health data such as blood pressure or heart rate, will be of particular importance to dementias research, where records of the behavioural changes over time that this data can reflect will provide deep and novel insights into the diseases.

Methods: The Wearables and Connected Devices Platform, hosted within the Health eResearch Centre at the University of Manchester and part of the wider Dementias Platform UK (DPUK) Informatics Platform, offers both a pool of wearable and mobile devices and the computational backend infrastructure that enables the scalable capturing of data from these devices. Designed to be utilised as a ready-made platform component by wider research projects in the dementias space the platform provides the technology, software and expertise needed to best take advantage of the potential that exists in this area.

Results: Currently the platform is providing wearables research infrastructure as part of the Deep and Frequent Phenotyping Study, which offers an exemplar use case of the platforms utility and functionality. In this presentation the wearables platform will be demonstrated alongside the range of devices that are currently being used.

Conclusions: The need for improved exposure and outcome measures grows. Wearables provide enormous potential to increase the precision and diversity of measures available to dementia research. The DPUK environment allows data from wearable and other smart devices to be integrated with other modality data including cohort data to establish more comprehensive datasets.