Using Video Software to Detect Atrial Fibrillation

By Savanna Ronco

       Patients with atrial fibrillation may one day be able to get a diagnosis of their irregular heartbeat using only a webcam, according to a new proof-of-concept study.

The study, done by scientists at the University of Rochester in New York, demonstrates that video monitoring software can detect when a person’s skin changes color due to changes in blood flow during AF (atrial fibrillation). Because of this, scientists were able to quickly identify those suffering with AF without ever having physical contact with the patient.

“We don’t need a lot of recording,” said researcher Dr. Jean-Philippe Couderc. “Just 15 seconds is valuable enough for us to detect the presence of atrial fibrillation in a patient.”

There were 11 patients included in the study who were already undergoing an electrocardiogram (ECG) test. They were recorded before, during and after the test by a webcam that was placed one meter in front of them. Researchers were then able to extract the patient’s pulse rate per minute.

After, they matched the video recording with the actual ECG test to determine if the video monitoring test was correct.

“It’s a simple concept,” Couderc said, “but one that enables more people with AF to get the care that they need.”

According to the study, the new video technology has a detection error rate of 20 percent, while the error rate of the ECG ranges between 17 and 29 percent.

AF is one of the most common heart rhythm problems. According to the study, which was published in Heart Rhythm in August 2014, it is estimated that 33.5 million people across the world have AF. An estimated 30 percent of those people don’t even realize their diagnosis.

Having AF means the upper chambers of the heart do not pump efficiently, which could potentially cause blood clots and raise the chance of stroke. A heart with AF may beat over 140 beats a minute, whereas a normal heart rate is between 60 and 100 beats a minute.

Couderc believes that this new technology demonstrates that AF can in fact be diagnosed in a way other than the traditional hooking a patient up to an ECG machine. Not only that, but it can also be diagnosed without ever needing to have contact with the patient.

This study is part of a larger collection of research based on using contactless video recordings for a broader range of diagnostics.

“When you think about this, the applications are enormous,” Couderc explained. “You could monitor people at home, at work or anywhere. It’s just a simple webcam.”

Though this study only included white patients, Couderc said that the pulse per minute could be detected in other ethnicities as well, including African Americans and Asians. However, one woman’s cosmetic foundation did block the observable variations in skin color. This could be a possible limitation to the software.

“This study was designed to be a proof of a concept,” Couderc said. “I think in terms of that it was a success. Hopefully we’ll be able to do great things with this technology in the future.”


One thought on “Using Video Software to Detect Atrial Fibrillation

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )


Connecting to %s