Researchers from the University of Utah and University of Illinois Chicago have developed a wearable device capable of continuously measuring blood pressure and blood flow without the need for a traditional inflatable cuff. Combining physics-based modelling with artificial intelligence, the technology could pave the way for more comfortable and comprehensive cardiovascular monitoring.
Blood pressure is one of the most important indicators of cardiovascular health, yet conventional monitoring relies on occasional readings taken with a cuff. While effective, these measurements provide only brief snapshots of a patient’s condition and often fail to capture fluctuations that occur throughout the day.
According to researcher Benjamin Sanchez Terrones, continuous blood pressure monitoring has long been considered a major challenge in healthcare technology. High blood pressure is often referred to as a “silent killer” because it significantly increases the risk of heart attacks, strokes and aneurysms. The newly developed smartwatch aims to address this limitation by continuously tracking blood pressure changes rather than recording isolated measurements. The findings were published in Nature Communications.
Electrical signals
Many wearable health devices estimate blood pressure using optical sensors that analyze changes in blood flow through light. However, the scientific basis of these methods is not always fully understood, and many systems rely heavily on AI models that operate as “black boxes,” making their outputs difficult to interpret clinically. The new device takes a different approach. Instead of light, it uses a painless and imperceptible electrical current to measure bioimpedance, the ease with which electricity travels through blood and surrounding tissues.
Because blood flow changes with every heartbeat, subtle variations in electrical conductivity contain information about both blood pressure and circulation. These signals are then processed by an AI model that incorporates established physical principles. According to co-author Christel Hohenegger, embedding physics directly into the model improves reliability because the system is constrained by real physiological processes rather than relying solely on statistical predictions.
Physics-informed AI
The researchers combined fluid dynamics, which describes how blood moves through vessels, with electromagnetism, which governs the bioimpedance measurements. This gives the system a transparent scientific foundation and helps prevent unrealistic predictions.
Unlike traditional blood pressure monitors that report only systolic and diastolic values, the new technology reconstructs the entire blood pressure waveform over time. According to co-author Braxton Osting, blood pressure should be viewed as a continuously changing physiological signal rather than a pair of static numbers.
Capturing the full waveform allows clinicians to observe how blood pressure responds to daily activities such as walking, exercising or climbing stairs, potentially providing richer insights into cardiovascular health. Another advantage is that the device does not require individual user calibration, a common limitation of many wearable blood pressure technologies.
Test results
To validate the technology, the researchers tested the device on approximately 150 individuals, including patients in intensive care units and outpatient clinics. This broader testing approach was designed to evaluate performance not only in healthy volunteers but also in patients with varying health conditions.
The results suggest that the smartwatch can provide continuous cardiovascular monitoring across different clinical environments. The University of Utah currently holds the intellectual property associated with the technology and is exploring opportunities to bring the innovation to market.
While further clinical validation will be needed before widespread adoption, the study highlights how the combination of artificial intelligence and physics-based modelling may transform blood pressure monitoring. By moving beyond intermittent measurements and capturing the complete “movie” of cardiovascular activity, the technology could support earlier detection of health problems and more personalized management of cardiovascular disease.
