A research team at the University of California San Francisco (UCSF) found that the Apple Watch could be a useful diagnosis tool for high blood pressure and sleep apnea if paired with an app developed by Cardiogram.
Cardiogram and UCSF's study on heart health is the third major study of deep learning in medicine, following Google Brain's results on diabetic eye disease in December of 2016 and a Stanford study on skin cancer this January.
The Apple watch used in the experiments was able to detect abnormal heart rhythm with an accuracy of 97 percent. Cardiogram's app called "DeepHeart" is connected to a neural network that can spot high blood pressure in 82% of cases and sleep apnea in 90% of cases.
To be clear, the Apple Watch doesn't have the kind of high-tech lab equipment usually used to detect these common yet serious conditions. An estimated 22 million adults in the USA are affected, with 80 percent of cases going undiagnosed, according to the American Sleep Apnea Association.
In the study, researchers looked at the heart rate data from more than 6,000 Apple Watch users.
It affects 22million adults in the USA and if left untreated the oxygen deprivation from sleep apnea can result in a growing number of health problems, including high blood pressure, stroke, heart failure, diabetes and heart attacks.
The research claims that data from wearable heart rate sensors, when combined with machine learning algorithms, can surface hidden patterns that predict whether a person is at risk for certain health problems.
The study authors wrote that the algorithm could offer a "surprisingly good prediction of hypertension and sleep apnea given that its only inputs are heart rate and step count". All of the participants used the Cardiogram app in their Apple watches over a period of time.
"The idea here is that by screening continuously you would identify people with hypertension who might not know they have it", said Johnson Hsieh, Co-Founder of Cardiogram.
He adds: 'Then you'd guide them through the appropriate final diagnosis, which would be through a blood pressure cuff and then treatment'.