Smartphones can be used to detect atrial fibrillation with existing hardware, according to research presented at the European Society of Cardiology’s annual meeting in Rome, featuring a low-cost application that uses the phone’s own accelerometer and gyroscope to check for atrial fibrillation.

“Atrial fibrillation is a dangerous medical condition present in 2% of the global population and accounting for up to seven million strokes per year,” said lead author Tero Koivisto, a vice-director of the Technology Research Centre, University of Turku, Finland. “In the European Union alone this heart rhythm disorder costs approximately $19 billion every year.”

Around 70% of strokes due to atrial fibrillation could be avoided with pre-emptive medication. But the condition often occurs randomly and is difficult to detect by visiting a doctor. There are relatively large and costly electrocardiogram devices that patients can take home for long-term monitoring, but they require a patch or wires that are clumsy to use and continuous contact with electrodes tends to irritate the skin.

Due to these constraints, current methods for detection of atrial fibrillation are infeasible for wide-scale screening of populations or higher risk age groups (60 years and above).

The study tested the ability of a smartphone to detect atrial fibrillation without any add-on hardware. The study included 16 patients with atrial fibrillation from the Turku Heart Centre. In addition, 20 recordings from healthy people were used as control group data to validate the developed algorithm.

To detect atrial fibrillation, a smartphone was placed on the chest of the patient, and accelerometer and gyroscope recordings were taken. Patients were advised to lie in a prone or supine position during the measurements.

“We use the accelerometer and gyroscope of the smartphone to acquire a heart signal from the patient,” said Koivisto. “A measurement recording is taken, and the acquired data is pre-processed by signal processing methods. Multiple features such as autocorrelation and spectral entropy are then extracted from the pre-processed data. Finally, a machine learning algorithm is used to determine if the patient suffers from atrial fibrillation.”

Using this technology the investigators detected atrial fibrillation with a sensitivity and specificity of more than 95%.

“We measure the actual motion of the heart through miniature accelerometers and gyroscopes that are already installed in today’s smartphones,” said Koivisto. “No additional hardware is needed and people just need to install an app with the algorithm we developed.”

European Society of Cardiology
escardio.org