Web cam used to detect abnormal heart rhythm

By Madeline Kennedy NEW YORK (Reuters Health) - With a regular laptop camera and sophisticated software, researchers may be able to detect atrial fibrillation about as accurately as with a standard electrocardiogram (ECG), according to a new pilot study. The technology records and analyzes video footage of a person’s face and detects subtle shifts in skin color that indicate changes in blood flow. “Existing ECG-based technologies are used only for patients who had symptoms,” said the Jean-Phillipe Couderc of the University of Rochester Medical Center in New York, who led the study. But there are “approximately 3.2 million people with AF in the U.S., and an estimated 30 million people in the world,” he told Reuters Health, and currently there’s no way to diagnose the condition without consulting a doctor. In atrial fibrillation (AF), the upper and lower chambers of the heart beat out of sync with each other. Although that can cause symptoms like heart palpitations or dizziness, often the condition has no signs. An estimated 30 percent of people with AF do not know they have it, according to past research. The long-term implications are serious, as AF is progressive and can lead to stroke or heart failure. Couderc said the video-based system, developed with colleagues at Xerox Research Center in Webster, New York, is the first diagnostic tool to measure the electrical activity of the heart without physical contact with skin. He envisions the face scan being used one day to diagnose patients without symptoms and monitor their progress as they receive treatment. For the pilot study, Couderc and his team enrolled 11 patients with diagnosed atrial fibrillation who were scheduled to undergo ECG at the university’s hospital. The researchers set up a standard web camera to capture video of the person’s face in 15-second increments at the same time as the ECG was being performed. The scan focuses on the face because its skin is particularly thin, making blood running through the veins more easily visible. In particular, the program looks for the color of the blood as it reflects or absorbs green light. The experiment demonstrated that the observed color changes corresponded with the subject’s heart rate, the study team reports in the journal Heart Rhythm. The scan correctly identified the presence or absence of atrial fibrillation 80 percent of the time, with an error rate of 20 percent. That was similar to the automated ECG measurements, which were wrong between 17 percent and 29 percent of the time. This new technology could one day be used by anyone with a tablet or laptop, Couderc said, though he cautioned that the small study is only a proof of concept and “we need to evaluate the technology in a large prospective clinical trial,” to test how accurately it works and how easy it is to use. Dr. Steven Swiryn of Northwestern University in Chicago, who studies atrial fibrillation, agreed that the small study is “quite preliminary, and though interesting, has a number of major limitations.” One, is that patients need to be absolutely still for the scan, an issue that excluded one participant in the study. Another participant was excluded because she was wearing makeup, which obstructed the camera’s ability to see blood flow. Finally, Swiryn noted, the study only included Caucasians and the technology needs to be tested on people of color as well. While it may not be possible to diagnose atrial fibrillation without a doctor, the condition can also “be strongly suspected by taking the pulse,” Swiryn said in an email. (See Reuters Health article of July 23, 2014, here: http://reut.rs/YqvCTe). However, Swiryn added, if the new technique’s methods were refined and applied to a larger and more diverse group, it would be worth pursuing with more research and “might prove a useful and inexpensive screening method.” SOURCE: bit.ly/W84G9h Heart Rhythm, online August 29, 2014.