SmartSocks: a new data collection paradigm for dementia and other neurological disorders

Simple item page
dc.contributor.authorSteer, Zeke
dc.contributor.authorVenkatesh, Prabha Thirthahalli
dc.contributor.authorMejia‐Mejia, Elisa
dc.contributor.authorDennis, William Wu
dc.contributor.authorOgundele, Patrick Ademola
dc.contributor.authorBrooking, Annie
dc.contributor.authorEimontaite, Iveta
dc.date.accessioned2025-01-27T15:07:50Z
dc.date.available2025-01-27T15:07:50Z
dc.date.freetoread2025-01-27
dc.date.issued2024-12
dc.date.pubOnline2025-01-09
dc.description.abstractBackground Distress and agitation are predictors of entry into long‐term care and health inequalities (Schulz et al., 2004, Weir et al., 2022). Physiological data has been shown to reliably predict distress (Goodwin et al., 2019), yet wearable devices have low acceptance rates (Koumpouros &amp; Kafazis, 2019). The current study discusses findings from a multifaceted approach investigating the detection of early signs of distress via physiological sensors in a foot‐worn device. Method Firstly, the acceptance and concern ratings for a foot‐worn device, SmartSocks, wrist‐worn devices, Empatica E4 and Shimmer GSR+, and chest‐worn device, Equivital within a healthy population (N = 10) were assessed with a self‐report questionnaire. Secondly, data accuracy between Shimmer ECG and Polar OH1+ was compared within a healthy population (N = 12) in a standing, sitting and supine position. Finally, an ongoing ecologically valid feasibility trial (N = 2) involving participants with dementia or a learning disability is assessing the reliability of physiological data and AI‐detected stress from SmartSocks relative to subjective ratings of distress, the Abbey Pain Scale (APS), and the Neuropsychiatric Inventory (NPI). Result Firstly, the SmartSocks received lowest concern ratings compared to wrist‐ and chest‐worn devices (1.64 vs <1.71). Secondly, the accuracy of SmartSocks pulse rate (PR) estimates obtained using photoplethysmography (PPG) in combination with the delineator algorithm was determined by comparing estimates to a Shimmer 1‐lead ECG, recording Mean Absolute Error (MAE)<5bpm at 64HZ for participants in a supine position (Fig. 1). This led to the development of new features for classifying PPG signal quality using neural networks, achieving approximately 95% accuracy. Finally, the initial stage of the feasibility trial indicated APS and NPI scores were lower after the participant with dementia wore SmartSocks for two weeks. Physiological data collected from the participant with a learning disability using SmartSocks showed moderate correlation (χ2 = 0.45) between the reported and AI‐detected stress over the day (Fig. 2 & 3). Conclusion Early findings suggest SmartSocks are more comfortable than comparable wrist‐ and chest‐worn devices, and validity of the data is comparable to other devices. Preliminary data obtained from people with dementia and learning disabilities suggest SmartSocks are capable of detecting distress to alleviate user discomfort.
dc.description.journalNameAlzheimer's & Dementia
dc.identifier.citationSteer Z, Venkatesh PT, Mejia‐Mejia E, et al., (2024) SmartSocks: a new data collection paradigm for dementia and other neurological disorders. Alzheimer's &amp; Dementia, Volume 20, Issue S8, December 2024, Article number e095065
dc.identifier.eissn1552-5279
dc.identifier.elementsID562442
dc.identifier.issn1552-5260
dc.identifier.issueNoS8
dc.identifier.paperNoe095065
dc.identifier.urihttps://doi.org/10.1002/alz.095065
dc.identifier.urihttps://dspace.lib.cranfield.ac.uk/handle/1826/23435
dc.identifier.volumeNo20
dc.languageEnglish
dc.language.isoen
dc.publisherWiley
dc.publisher.urihttps://alz-journals.onlinelibrary.wiley.com/doi/10.1002/alz.095065
dc.rightsAttribution 4.0 Internationalen
dc.rights.urihttp://creativecommons.org/licenses/by/4.0/
dc.subject32 Biomedical and Clinical Sciences
dc.subject5202 Biological Psychology
dc.subject3202 Clinical Sciences
dc.subject3209 Neurosciences
dc.subject52 Psychology
dc.subjectClinical Research
dc.subjectAging
dc.subjectMachine Learning and Artificial Intelligence
dc.subjectMental Health
dc.subjectBioengineering
dc.subjectNeurosciences
dc.subjectBehavioral and Social Science
dc.subjectNeurological
dc.subjectGeriatrics
dc.subject3202 Clinical sciences
dc.subject3209 Neurosciences
dc.subject5202 Biological psychology
dc.titleSmartSocks: a new data collection paradigm for dementia and other neurological disorders
dc.typeArticle

Files

Original bundle
Now showing 1 - 1 of 1
Thumbnail Image
Name:
SmartSocks-2024.pdf
Size:
8 MB
Format:
Adobe Portable Document Format
Download
License bundle
Now showing 1 - 1 of 1
No Thumbnail Available
Name:
license.txt
Size:
1.63 KB
Format:
Plain Text
Description:
Download

Collections

Staff publications (SATM)