‘Radar stethoscope’ improves contactless health technology

Researchers from the University of Glasgow led the development of the new system, which uses radar to track patients’ heart sounds like a doctor uses a stethoscope.  It improves significantly on previous methods of measuring heart rate using radar waves, which take readings from measurements of patients’ chest movements.

In a new paper published in the journal IEEE Journal of Biomedical and Health Informatics, the team demonstrate how they used advanced signal processing techniques to isolate and measure the heart sounds of human volunteers.

Using a 24Ghz continuous-wave radar system, they bounced electromagnetic waves off volunteers’ bodies as they lay down, fully-clothed. The reflections of the waves allowed the team to measure not just the movement of their chests but also the sounds their hearts produced as valves opened and closed – a method similar to the way doctors use stethoscopes to hear how patients’ hearts are beating.

They used sophisticated filters to remove signal noise and other interference, allowing them to get a clear pulse signal and calculate patients’ heart rate.

To test the effectiveness of their method, they collected heart sound and chest movement data from male and female volunteers using radar over periods of 30 seconds, 60 seconds, and five minutes. For each time period, they also measured three different intensities of heartbeat – a ‘resting’ state of between 60 and 80 beats per minute, an ‘anxiety’ state of 100-130 bpm, and a ‘transition’ state of between 80 and 130 bpm.

At the same time, they also monitored the volunteers’ hearts using an electrocardiogram (ECG) machine – the current ‘gold standard’ for clinical measurements of heart rate.

The researchers found that their new system was capable of measuring heart sounds across all intensities with nearly 99% accuracy – their readings differed from the simultaneous ECG measurements by less than one beat per minute.

They also compared the data they collected on chest movement – the conventional method of measuring heart rate with radar – to their improved heart sound method.

They found that the chest movement measurements could fall short of the ECG measurements by between eight and 50 beats depending on the intensity of the volunteers’ heart beat, highlighting the significant improvement their new method represents.

Professor Qammer H. Abbasi, co-head of the Communications, Sensing and Imaging Hub at the James Watt School of Engineering is one of the paper’s contact authors. He said: “In recent years, advances in sensing technologies have opened up promising new methods of tracking patients’ vital signs without physical contact. These advances could help provide clinicians with round-the-clock monitoring of their patients without the use of invasive or uncomfortable wearable sensors, and strengthen infection control measures in cases where patients have a communicable illness.

“What we’ve been able to with this research is take a big step towards enabling the full potential of radar as a contact-free health monitoring tool. Our signal processing approach has enabled much more accurate detection of heart sounds and measurement of heart rate, which bring it close to the performance of ECG monitoring.”

Professor Muhammad Imran is head of the University of Glasgow’s Communications, Sensing and Imaging Hub, and a co-author of the paper. He said: “This paper shows that radar can be used to monitor heart sounds with remarkable precision, which could make it invaluable for use in clinical settings and at home in the future. We’re already looking at other ways to precisely read people’s other vital cardiovascular signs using this technique. We hope to develop a more fully-featured commercial design in the future which could pair heart rate monitoring with breathing rate, blood pressure readings and other useful measurements.”

Researchers from Coventry University also contributed to the paper. The team’s paper, titled ‘Contactless Heart Sound detection using Advanced Signal Processing Exploiting Radar Signals’, is published in the IEEE Journal of Biomedical and Health Informatics.