Real-world deployment of remote sleep monitoring technologies reveals distinct patterns associated with cognitive decline
File(s)
Published version (article in press)
Author(s)
Fletcher-Lloyd, Nan Victoria
Céspedes Gómez, Nathalia
Capstick, Alexander
Fogel, Antigone
Bafaloukou, Marirena
Type
Journal Article
Abstract
Examining sleep patterns in relation to chronological ageing and dementia can provide insights for risk screening. Integrating predictive models with remote sleep monitoring enables routine assessment of cognitive decline symptoms in high-risk groups,
aiding early risk identification. We developed a machine learning pipeline to estimate Sleep Age Index from longitudinal under the-mattress sleep sensor data in the general population and a dementia cohort (n=1,672; person-samples=18,369), using it to identify dementia risk. Risk scores were stratified into high, medium and low-risk categories to support clinical decision-making.
Our study indicates that sleep patterns in dementia do not follow typical ageing processes, with the pre-trained model showing greater deviation from ”normative” age-related patterns. These deviations were associated with irregular bed- and rise-times and reduced night-to-night variability in deep sleep. Chronological age was predicted from sleep data with a mean absolute error
of 5.52 (95% CI: 5.37-5.67) on held-out data. In dementia versus control, the model achieved 75.7% (95% CI: 71.4%-79.9%) sensitivity and 74.7% (95% CI: 69.2%-80.0%) specificity post-stratification on unseen data. In a pilot high-risk cohort (n=50), model predictions showed slight positive bias relative to clinical judgement (mean difference 0.98, limits of agreement-0.83
2.78). These findings demonstrate the potential of remote sleep monitoring and predictive modelling in identifying individuals who may benefit from further clinical evaluation and early intervention.
aiding early risk identification. We developed a machine learning pipeline to estimate Sleep Age Index from longitudinal under the-mattress sleep sensor data in the general population and a dementia cohort (n=1,672; person-samples=18,369), using it to identify dementia risk. Risk scores were stratified into high, medium and low-risk categories to support clinical decision-making.
Our study indicates that sleep patterns in dementia do not follow typical ageing processes, with the pre-trained model showing greater deviation from ”normative” age-related patterns. These deviations were associated with irregular bed- and rise-times and reduced night-to-night variability in deep sleep. Chronological age was predicted from sleep data with a mean absolute error
of 5.52 (95% CI: 5.37-5.67) on held-out data. In dementia versus control, the model achieved 75.7% (95% CI: 71.4%-79.9%) sensitivity and 74.7% (95% CI: 69.2%-80.0%) specificity post-stratification on unseen data. In a pilot high-risk cohort (n=50), model predictions showed slight positive bias relative to clinical judgement (mean difference 0.98, limits of agreement-0.83
2.78). These findings demonstrate the potential of remote sleep monitoring and predictive modelling in identifying individuals who may benefit from further clinical evaluation and early intervention.
Date Issued
2026-07-21
Date Acceptance
2026-06-26
Citation
npj Digital Medicine, 2026
ISSN
2398-6352
Publisher
Nature Portfolio
Journal / Book Title
npj Digital Medicine
Copyright Statement
© The Author(s) 2026. Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/.
License URL
Publication Status
Published online
Date Publish Online
2026-07-21