Post

New research · AI in Medicine
NPJ digital medicine · 1d
Cohort studyNPJ digital medicine · 2026

Real-world deployment of remote sleep monitoring technologies reveals distinct patterns associated with cognitive decline.

Nan Fletcher-Lloyd, Nathalia Céspedes Gómez, Alexander Capstick … Payam Barnaghi
Read paper
AI in MedicineCohort study

Remote sleep monitoring model identifies dementia with 75.7% sensitivity.

Real-world deployment of remote sleep monitoring technologies reveals distinct patterns associated with cognitive decline.

Nan Fletcher-Lloyd et al. · NPJ digital medicine · 2026
Background

Examining sleep patterns in relation to chronological ageing and dementia can provide insights for risk screening.

Methods

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 = 1672; person-samples = 18,369), using it to identify dementia risk.

n = 1672
Results

This is the percentage of people with dementia that the sleep model correctly identified

mean difference 0.98 (95% CI -0.83 to 2.78)
null = 0-0.832.78
CI crosses the null - not significant
More results

Integrating predictive models with remote sleep monitoring enables routine assessment of cognitive decline symptoms in high-risk groups, aiding early risk identification.

Risk scores were stratified into high, medium and low-risk categories to support clinical decision-making.

More results

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 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).

“
Conclusion

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.

Read paper
0 comments

No comments yet. Be the first.

Related papers

LatestFoundational
AI / Informatics
27,624 words
total Simplified Chinese words with new AI familiarity estimates now available
AI / Informatics
89%
Artificial intelligence/machine learning models correctly identified 89% of high-grade glioma recurrences and non-recurrences.
Systematic Review
“Our review suggests that LDCT screening is generally cost-effective in high-risk populations, although implementing LC screening programs may entail a substantial budgetary impact.”
Guideline
“By outlining key strategic, regulatory, and operational considerations, this primer aims to equip clinician innovators with foundational knowledge to navigate the pathway from unmet clinical need to scalable endoscopic technology.”
AI / Informatics
0.761
how well the artificial intelligence system predicted how long patients with pancreatic cancer would live
AI / Informatics
C-index = 96.0%
The model accurately predicted when initial leukemia treatment would not work
AI / Informatics
0.64-0.69
The modest ability of AI-ECG alone to predict irregular heartbeat risk
AI / Informatics
0.987
model's accuracy in identifying Cadmium contamination levels