Not a very smart watch
Are smartwatches good enough to detect atrial fibrillation?
Julia Tisler, ERR
Smart watches and fitness bracelets are becoming more and more sophisticated, but it is too early to rely on their results of measuring health indicators. Scientists have found that mobile applications for detecting atrial fibrillation often give false positive and inconclusive results in some patients with certain heart diseases.
Advanced cardiac monitoring of patients and the use of implantable cardiovascular electronic devices may increase the likelihood of detecting atrial fibrillation, but these devices have limitations, including short battery life and lack of immediate feedback.
Therefore, scientists wondered if it was possible to use smartphone apps and smart watches for monitoring. The largest study to date published in The Canadian Journal of Cardiology (Racine et al., Role of Coexisting ECG Anomalies in the Accuracy of Smartwatch ECG Detection of Atrial Fibrillation), shows that the use of these devices in patients with abnormal ECG (electrocardiography) is difficult. However, scientists say that better algorithms and machine learning can help these tools make more accurate diagnoses.
"Previous studies have confirmed the accuracy of the Apple Watch for the diagnosis of atrial fibrillation in a limited number of patients with similar clinical profiles," explained lead researcher Marc Strick, MD from the University Hospital of Bordeaux (France).
The study included 734 hospitalized patients. Everyone was given an ECG, after which the indicators were immediately checked using the app on the Apple Watch. It turned out that in about one in five patients, measurements made by a smartwatch did not allow for an automatic diagnosis. The risk of false-positive automatic detection of atrial fibrillation was higher in patients with premature atrial and ventricular contractions (PSG), sinus node dysfunction and atrioventricular block of the second or third degree. For people with arrhythmia, the risk of false negative recording was higher if they had ventricular conduction disorders (delayed ventricular conduction).
The smartphone app correctly identified 78% of patients with atrial fibrillation and 81% of patients without this disease. Electrophysiologists identified 97% of patients with arrhythmia and 89% without it.
Patients with PSG were three times more likely to have a false positive diagnosis based on smartwatch data, and the identification of patients with atrial tachycardia and atrial flutter was very poor.
"These observations are not surprising, since the algorithms for automatic detection of smartwatches are based solely on the variability of the cycle," said Dr. Strick.
In the accompanying article , Doctor of Medicine Andres F. Miranda-Arboleda and Adrian Baranchuk, MD, from the Department of Cardiology at the Kingston Health Science Center (Canada), noted that this is the first study devoted to the use of the Apple Watch as a diagnostic tool for atrial fibrillation.
"In a certain sense, smartwatch algorithms for detecting arrhythmias in patients with cardiovascular diseases are not smart enough yet. But they may get better soon," said Dr. Miranda—Arboleda and Dr. Baranchuk.
Portal "Eternal youth" http://vechnayamolodost.ru