New Algorithm Measures Heart Rate by Head Movements

Print


A newly developed algorithm has the ability to measure heart rates of individuals displayed on a digital video-platform by observing indiscernible head movements that come with a rush of blood due to the heart’s activity.

Researchers from the Massachusetts Institute of Technology (MIT) Computer Science and Artificial Intelligence Laboratory Department constructed the algorithm that issued pulse estimations that kept pace within a few beats per minute of those released by electrocardiograms (EKGs).

Furthermore, the algorithm was shown to be able to deliver a rough calculation of the time gaps between beats, a method used to identify patients who may be at cardiac risk.
The team of researches is set to reveal their findings this summer at the Institute of Electrical and Electronics Engineers' Computer Vision and Pattern Recognition conference.

This new digital video-platform, pulse-calculating algorithm could be utilized in time to check the heart rates of senior citizens or newborns, without running the risk of repeatedly using EKG leads that may irritate or harm sensitive skin.Algorithm head movement

"From a medical perspective, I think that the long-term utility is going to be in applications beyond just pulse measurement. Can you use the same type of techniques to look for bilateral asymmetries? What would it mean if you had more motion on one side than the other?" said Dugald C. Jackson Professor of Electrical Engineering and Computer Science and director of MIT's Data-Driven Medicine Group, John Guttag.

The algorithm is a blend of different techniques known to the field of computer vision. It begins by using basic face recognition to distinguish between a patient’s head from the rest of the image. Then it arbitrarily selects 500 to 1,000 precise points, cluttered around the patient’s nose and mouth, following every movement frame by frame.

From there, it sifts out any frame by frame movements that lose its overall frequency/detection levels outside the range of a standard heart beat, about 0.5 to 5 hertz, or 30 to 300 cycles per minute. This will discard any movements that persist at a lower frequency, such as those caused by normal breathing and slow changes in stance.

Finally, using a technique known as principal component analysis, the algorithm breaks down the resulting signal into numerous component signals, which stand as part of the remaining movements that have no association with one another. From those signals, it selects one that seems to be the most regular and that drops within the usual frequency band of the human pulse.

Guttag went on to state that theoretically, the system could determine cardiac output, or the quantity of blood pumped by the heart, which is often a means of diagnosis in most cardiac cases.

“Before the echocardiogram, cardio output was measured by calculating exactly the types of mechanical forces that the new algorithm registers. I think this should be viewed as proof of concept. It opens up a lot of potential flexibility,” he said.