Cardiovascular digital biomarkers
Heart rate and rhythm are the signals wearables capture best, and continuous capture turns a population comparison into a personal baseline.
%20(1).png)
The cardiovascular system produces signals that wearables are unusually well suited to capture. Heart rate can be estimated optically from the wrist continuously and cheaply, rhythm can be recorded through single lead electrocardiography on consumer hardware, and both change in ways that carry clinical meaning long before a person notices symptoms.
This category collects the measures derived from those signals: resting and nocturnal heart rate, variability between beats, rhythm and arrhythmia detection, and the vascular measures that sit alongside them. Some are longstanding clinical parameters that wearables now capture continuously rather than at a single clinic visit. Others exist only because continuous recording became possible, such as burden measures that express how much of a period a person spent in an abnormal rhythm.
Continuous capture changes the questions that can be asked. A resting heart rate taken in clinic reflects one moment, including the effect of having travelled to the appointment. The same measure taken nightly across a month reveals a personal baseline and any drift away from it. Much of the clinical value in this domain comes from that shift, from comparing a person to a population norm towards comparing them to themselves.
Cardiovascular measures in this library
How these measures are used
In trials these measures appear as safety signals, as covariates and, increasingly, as endpoints in their own right. Heart rate is collected in almost every wearable study whether or not it is the focus, because it is needed to interpret activity data and to detect physiological stress. Rhythm measures are used in screening studies, in follow up after ablation or cardioversion, and in trials of anticoagulation strategy where the amount of arrhythmia matters.
Remote monitoring is the other major use. Nocturnal heart rate and rhythm burden can be reviewed between visits, which suits conditions where deterioration is gradual and episodic. Studies commonly pair these measures with a patient reported instrument covering symptom burden and quality of life, because a rhythm abnormality detected by a sensor and a symptom experienced by a person are related but not interchangeable, and treatment decisions usually need both.
What the evidence supports today
This domain contains both the strongest and some of the weakest evidence in the library, and the difference matters. Single lead electrocardiography on consumer wearables has cleared regulatory review in several markets, and the atrial fibrillation features built on it have accumulated large prospective screening studies. Optical heart rate at rest is well validated against chest strap electrocardiography. Those are solid foundations.
Heart rate variability is a different case. It is widely reported by consumer devices and widely used in commercial wellbeing products, but its appearance as a registered clinical trial endpoint is far rarer than its popularity suggests, and different devices compute it over different windows using different algorithms, which makes values incomparable. Blood pressure from a cuffless wearable remains an area where marketing has moved ahead of clearance.
The honest summary is that rhythm detection is mature, resting heart rate is dependable, and everything downstream of those two should be treated as promising rather than established.
Common questions
Can a smartwatch diagnose a heart condition?
No. Consumer electrocardiography features are cleared to detect specific patterns and to prompt a person to seek assessment, not to diagnose. A recording that suggests an abnormal rhythm still requires confirmation by a clinician, usually with a conventional twelve lead electrocardiogram or extended monitoring.
Is optical heart rate from the wrist reliable?
At rest and during sleep it agrees closely with chest strap electrocardiography. Accuracy falls during vigorous or irregular movement, in people with darker skin tone under some sensor configurations, and when the device is worn loosely. Studies that depend on exercise heart rate usually supplement wrist optical sensing with a chest strap.
Why is heart rate variability treated cautiously here?
Because the number depends heavily on how it was produced. Different devices sample over different windows, apply different artefact filters and report different metrics under the same label. Within one device and one person, trends can be informative. Across devices or between people, the comparison is usually not meaningful.
What is rhythm burden and why does it matter?
Burden expresses how much of a monitored period a person spent in an abnormal rhythm, rather than whether an episode was ever detected. It is a more informative endpoint than a yes or no detection because it can increase or decrease with treatment, which makes it usable as a response measure rather than only a diagnostic one.
Run a study on these measures
WeGuide captures wearable data and patient reported outcomes in one platform, from screening through to analysis.
