Cardiovascular Sensing Moves Beyond Wearables

For decades, monitoring cardiovascular health has required some combination of cuffs, sticky electrodes, wearable sensors, and frequent doctor visits. A new experimental system suggests the future of heart monitoring may look less like a hospital, and more like a smart speaker sitting quietly on a shelf at home.

Researchers from Carnegie Mellon University’s College of Engineering have developed an AI-powered millimeter-wave radar platform capable of tracking blood-flow dynamics across the human body without making physical contact. The prototype system, called PolyPulse, uses the same class of radar technology found in autonomous vehicles and consumer electronics to measure subtle bodily movements caused by cardiac activity. The result is a compact device that can estimate pulse transit time, a key marker of arterial stiffness, and even infer diastolic blood pressure remotely.

“The implications are potentially enormous,” says Jiangyifei Zhu, a Ph.D. student in electrical and computer engineering and co-lead author of the paper. “Cardiovascular disease remains the leading cause of death globally, and many of its warning signs develop gradually over years. By detecting these changes early at home could dramatically improve outcomes.”

Instead of attaching sensors to the skin, the system emits millimeter-wave “chirps” toward a seated individual and analyzes the reflected signals using beamforming techniques and neural networks. These reflections capture microscopic movements at several physiological locations simultaneously: the apex of the heart, the carotid artery in the neck, the radial artery in the wrist, and the mastoid region near the head. From these measurements, AI reconstructs pulse transit times along multiple vascular pathways. That matters because pulse transit time, the interval required for a pressure wave to travel through the arteries, acts as a proxy for arterial stiffness. In general, stiffer arteries transmit pressure waves more rapidly, a phenomenon associated with hypertension, stroke risk, coronary artery disease, and neurodegenerative conditions such as Alzheimer’s disease.

Recently published in Nature Communications, the research represents the culmination of an ongoing effort that began after the team received the competitive Qualcomm Innovation Fellowship award in 2025.

“Most previous contactless systems could only monitor a single pathway,” explains Kuang Yuan, a Ph.D. student in electrical and computer engineering and co-lead author of the paper. “PolyPulse instead captures several simultaneously, creating a more distributed map of cardiovascular behavior across the upper body.”

The research suggests this multi-site approach could reveal subtler physiological changes that single-point measurements could miss. Current camera-based health sensors can fail in low light or when clothing obscures the body. Radar, by contrast, operates independently of visible light and can detect motion through certain materials. That robustness makes it especially attractive for passive, long-term monitoring inside homes.

“We envision a future where cardiovascular sensing becomes ambient, integrated into the environment rather than attached to the body,” says Swarun Kumar, the Sathaye Family Foundation Professor of Electrical and Computer Engineering, co-author of the paper, and advisor to Yuan. “A person could potentially receive continuous monitoring while simply sitting at home, reading, or watching television.”

The hardware itself is surprisingly compact: roughly 16.5x14 centimeters, it’s small enough to integrate into future smart-home devices.

This research represents an early proof of concept rather than a clinical product. The team still needs to validate the system across larger and more diverse populations. Regulatory approval, medical calibration standards, and long-term reliability testing would all be required before physicians could rely on such systems diagnostically.

“This is still an early-stage proof of concept, but it demonstrates how AI and wireless sensing may eventually transform the home into a proactive health-monitoring space,” says Justin Chan, assistant professor of electrical and computer engineering and computer science, co-author of the paper, and co-advisor to Zhu.

Still, the vision is compelling: cardiovascular monitoring that requires no cuffs, no wearables, and no active participation from the patient. Just a quiet radar device, sitting unobtrusively in the corner of a room, listening to the body through reflected waves invisible to the human eye.

Authors of the paper include Jiangyifei Zhu, Kuang Yuan, Akarsh Prabhakara from the Department of Computer Sciences at the University of Wisconsin-Madison, Yunzhi Li and Gongwei Wang from the Department of Electrical and Computer Engineering at Carnegie Mellon University, Kelly Michaelsen from the Department of Anesthesiology and Pain Medicine at the University of Washington, Justin Chan, and Swarun Kumar.