The results of a study conducted by heartbeat measurement app Cardiogram and the University of California, San Francisco, have confirmed that the Apple Watch is as much as 97% accurate in distinguishing between a normal heartbeat and the common-most heart rhythm abnormalities when paired with an AI-based algorithm.
6,158 participants became a part of the study after a recruitment process conducted by Cardiogram app on Apple Watch. Majority of the participants in the UCSF Health eHeart study had normal EKG readings. But as many as 200 among them had previously been diagnosed with paroxysmal atrial fibrillation (an abnormal heartbeat). Engineers then utilized the neural networks to identify all these abnormal heart rhythms from Apple Watch heart rate data.
According to Brandon Ballinger, who is Cardiogram’s co-founder and data scientist for UCSF’s eHeart study, about 25% of strokes are the result of an abnormal heart rhythm, and the sole purpose of Cardiogram and UCSF, since the beginning of the test in 2016, has been to deduce whether the Apple Watch could detect an oncoming stroke.
Cardiograms deep neural networks was tested against 51 in-hospital cardioversions– a process for restoring hearts normal rhythm– and this confirmed that it achieved 97% accuracy through the neural networks in detecting irregular heart rhythms.
For now, the study is only pillared on on a preliminary algorithm, however, it displays a great potential of detecting and eliminating strokes in future. A quarter of all the strokes are virtually caused by Atrial fibrillation, which is the most common abnormal heart rhythm. Ballinger says that two-thirds of such strokes are preventable by some inexpensive drugs.
It is for this reason that more and more people, especially the elderly ones who are more prone to such harm, are switching to wearable tech like Fitbit or the Apple Watch, which can perfectly monitor the heart. Pairing them up with algorithms capable of detecting heart problems, will be helpful in saving too many lives.
Meanwhile, Cardiogram and UCSF will continue their venture in eHealth study and proving its deep neural network “against multiple gold standards, incorporating the results into the Cardiogram app itself, and investigating the ability to detect health conditions beyond atrial fibrillation”.
