Anemia is a common condition, but it often goes undiagnosed because confirming it requires a venous blood test. This study looks at whether a commercially available wrist-worn smartwatch can help identify people who may have anemia, without taking blood.
The watch continuously records several types of physiological signals, including photoplethysmography (a light-based measure of blood flow), movement, heart rate, blood oxygen saturation, and heart rate variability. Researchers will use these signals to build a computer model that sorts participants into two groups: likely to have anemia, or unlikely to have anemia.
About 400 adults between 18 and 80 years of age will take part at one hospital in Beijing, China. Both patients with anemia and people with normal hemoglobin levels will be included. Each participant will wear the watch and also have a standard venous blood test to measure hemoglobin, which serves as the reference for comparison.
Data from the first 250 participants will be used to develop the model. Data from the next 150 participants will be used to test how well the model works in a separate group. The main question is how accurately the watch-based result matches the blood test result.
Eligibility
Sex
ALL
Min age
18 Years
Max age
80 Years
Healthy volunteers
Accepted
Inclusion Criteria:
1. Age 18 to 80 years, any sex
2. Willing to participate and able to provide written informed consent
3. Able to comply with wearable device wear, venous blood sampling, and functional assessment (e.g., six-minute walk test)
4. For participants with anemia: a hemoglobin test result obtained within 3 days before or after wearable signal acquisition and a hospital diagnostic report, including anemia severity, red blood cell size classification, and anemia type. Anemia is defined according to Chinese sea-level criteria: hemoglobin below 120 g/L in adult men, below 110 g/L in non-pregnant adult women, and below 100 g/L in pregnant women
5. For control participants: hemoglobin within the normal reference range -
Exclusion Criteria:
1. Axillary temperature 37.3°C or higher on the day of data collection or on the preceding day
2. Acute exacerbation of a severe comorbid condition (e.g., decompensated heart failure, respiratory failure, acute infection)
3. Skin breakdown, rash, deformity, amputation, or any other condition of the upper limb precluding proper wear of the device
4. Refusal of, or intolerance to, invasive venous blood sampling
5. Concurrent participation in another interventional clinical trial -
Primary outcome measure(s)
Sensitivity and specificity of the wearable-based deep learning model for binary anemia classification — Day 1 Agreement between the model-derived anemia versus non-anemia classification and venous hemoglobin measurement as the reference standard. Venous blood sampling is performed within 3 days before or after wearable signal acquisition.Reported as sensitivity and specificity with 95% confidence intervals in the independent validation cohort.
Trial sites (1)
Facility
City
Region
Status
Peking Union Medical College Hospital
Beijing
Beijing Municipality
More Peking Union Medical College Hospital trials in China
This page summarises publicly available registry data for informational purposes — not medical advice. Eligibility is determined by each study team; patients should discuss participation with their clinician.
We use cookies to analyse site traffic and improve your experience. With your consent, we may also use cookies for advertising. You can change your choice at any time on our Cookie Policy page. See also our Privacy Policy.