Smartwatch used to detect irregular heartbeat, JAMA reports
A smartwatch coupled with a machine learning algorithm was able to detect irregular heartbeat, or atrial fibrillation, with high accuracy in a small group of patients undergoing treatment to restore normal heart rhythm, according to research published last week in JAMA Cardiology.
As many as 9,750 participants with an Apple Watch smartwatch enrolled in the Health eHeart Study, including 347 with self-reported AF, and another group of 51 patients undergoing cardioversion, a treatment using medication or electricity, to restore regular heart rhythm from 2016 to March 2017; participants wore smartwatches to collect heart rate and step count data as part of the development and training of a deep neural network, which is a type of machine learning algorithm, to detect AF.
"These data support further research regarding the use of commercially available smartwatches coupled with a deep neural network for the purpose of detecting AF," noted Gregory Marcus, a researcher with the University of California, San Francisco.
"As sensor technologies have miniaturized in size and cost, their penetration into the consumer wellness and retail space has intensified," wrote Mintu Turakhia, associate professor of Medicine (Cardiovascular Medicine) at the Palo Alto Veterans Affairs Health Care System, in an accompanying editorial. "Although most of these devices have not been integrated into routine clinical use for various reasons, they remain attractive targets for health care because of their potential to more easily access large populations for disease screening and management. Connectivity of these sensor devices to mobile phones, which are globally ubiquitous, simplify data collection at scale. At the same time, indefinite continuous ECG recording with wearables has been difficult because of issues of lead placement, electrode contact and battery drain. Heart rate sensors on watches and fitness bands use photoplethysmography — the digital version of pulse recordings first described more than a century ago. Therefore, an obvious question is whether these can be leveraged to detect arrhythmias."
AF detection was associated with a lower accuracy in a larger group of people with a self-reported history of AF.
Atrial fibrillation affects 34 million people worldwide and is a leading cause of stroke. AF often has no symptoms and it can go undetected until a stroke happens.
As many as 9,750 participants with an Apple Watch smartwatch enrolled in the Health eHeart Study, including 347 with self-reported AF, and another group of 51 patients undergoing cardioversion, a treatment using medication or electricity, to restore regular heart rhythm from 2016 to March 2017; participants wore smartwatches to collect heart rate and step count data as part of the development and training of a deep neural network, which is a type of machine learning algorithm, to detect AF.
"These data support further research regarding the use of commercially available smartwatches coupled with a deep neural network for the purpose of detecting AF," noted Gregory Marcus, a researcher with the University of California, San Francisco.
"As sensor technologies have miniaturized in size and cost, their penetration into the consumer wellness and retail space has intensified," wrote Mintu Turakhia, associate professor of Medicine (Cardiovascular Medicine) at the Palo Alto Veterans Affairs Health Care System, in an accompanying editorial. "Although most of these devices have not been integrated into routine clinical use for various reasons, they remain attractive targets for health care because of their potential to more easily access large populations for disease screening and management. Connectivity of these sensor devices to mobile phones, which are globally ubiquitous, simplify data collection at scale. At the same time, indefinite continuous ECG recording with wearables has been difficult because of issues of lead placement, electrode contact and battery drain. Heart rate sensors on watches and fitness bands use photoplethysmography — the digital version of pulse recordings first described more than a century ago. Therefore, an obvious question is whether these can be leveraged to detect arrhythmias."
AF detection was associated with a lower accuracy in a larger group of people with a self-reported history of AF.
Atrial fibrillation affects 34 million people worldwide and is a leading cause of stroke. AF often has no symptoms and it can go undetected until a stroke happens.