DeepHeart AI IDs sleep apnea, hypertension via Apple Watch

Your Apple Watch can tell if you have hypertension or sleep apnea — with the help of Cardiogram’s deep neural network, DeepHeart, that is. The app-maker and the UCSF Health lab have conducted a study proving that wearables can suggest the presence of hypertension and sleep apnea with 82 percent and 90 percent accuracy, so long as they come equipped with heart rate sensors and accelerometers.

And, yes, they’re not just talking about Apple Watch, but also Android Wear devices, Garmin and Fitbits.


According to Cardiogram, heart rate sensors can detect both conditions, because your body’s autonomic nervous system connects your heart with the brain, stomach, esophagus, liver, intestines, pancreas and blood vessels. The company needed data to be able to train its AI to recognize heart rate patterns that denote the presence of the conditions, though, so it recruited 6,115 of its app’s users to participate in an online study with the UCSF Health lab.


Technology That Helps
Your Apple Watch can tell if you have hypertension or sleep apnea.


Cardiogram managed to collect 30 billion sensor measurements composed of the subjects’ heart rate and step counts, which is then fed to its AI. Since 37 percent and 17 percent of the participants have hypertension and sleep apnea, respectively, the AI got the samples it needed to be able to say whether someone has either condition.

According to the World Health Organization, 1 in 5 people with hypertension are undiagnosed — perhaps even more now that the American Heart Association has redefined the condition’s treatment guidelines. At the same time, 80 percent of people with sleep apnea — a disorder that disrupts your breathing during sleep — don’t even realize they have it. A wearable that can detect them could compel you to seek medical attention and ultimately save your life.

The app makers studies still have to undergo peer-reviewed clinical research, and that includes its previous experiment, wherein it trained its AI to detect the signs of a stroke. It promises to “start translating these research results into actual care” these next few months, as well as to expand its AI’s abilities and give it the power to detect diabetes.



Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

%d bloggers like this: