Daily rhythm and metabolic syndrome: a study of 272 adults
Study of 272 adults with a wrist device: metabolic syndrome was linked more to heart rate rhythms than to sleep. An association, not a cause.
Blog · Sleep
Study of an algorithm trained on age and nine sleep measures recorded at home: it matched the diagnosis in over 75% of cases. It still needs validation.
Updated October 20, 2026
A study in Scientific Reports (2024) trained a four-class machine learning algorithm on age and nine sleep biomarkers. The classes were Alzheimer's disease dementia, Lewy body dementia, prodromal synucleinopathy (people with isolated REM sleep behaviour disorder) and controls without neurodegenerative disease. The algorithm was then tested on 381 recordings, which also included people with Parkinson's disease without dementia and with mild cognitive impairment.
The nine biomarkers are REM sleep time, sleep spindle duration, atypical N3 sleep, non-REM hypertonia, autonomic activation, non-REM relative theta, non-REM theta/alpha ratio, sleep efficiency and supine sleep. They were recorded at home with a dedicated device that picks up frontal EEG, eye movements, pulse, head position and snoring: these are not measures an ordinary ring or watch can provide.
Lewy body dementia can be hard to tell apart from Alzheimer's and Parkinson's. In the study, the sleep profile matched the clinical diagnosis in over 75% of cases for Alzheimer's, Lewy body dementia and controls. On a repeat recording, the profile was the same or similar in 88% of people with Lewy body dementia, 86% of those with Alzheimer's and 98% of controls.
According to the authors, biomarkers for early detection of neurodegeneration are needed to treat symptoms earlier and to identify people who may benefit from neuroprotective interventions. They also write that the method still needs development and validation in larger groups.
Sleep duration and sleep regularity are not the same thing.
The study used a dedicated device and a clinical diagnosis: this is not what an app does. The Digital Twin lines up the data your devices write to your phone, such as steps, heart rate and sleep, and shows you how they change over time. It does not compare your sleep with this algorithm's profiles, it does not flag disease and it does not diagnose: this research describes groups of people, not you. A neurological assessment is for your doctor to decide.
The source is at the bottom. Open it.
This summary starts from research by others. It is not medical advice for you: talk it over with your doctor.
Study of 272 adults with a wrist device: metabolic syndrome was linked more to heart rate rhythms than to sleep. An association, not a cause.
Study of 413 people with early Parkinson's: probable REM sleep behaviour disorder was linked to faster cognitive decline. An association, not a cause.
Study of 1,992 adults: highly variable sleep duration and timing were linked to about double the cardiovascular risk. An association, not a cause.
More studies, same rule: read the source.