Cardiogram raises $2m to foretell atrial fibrillation using Apple Watch

Cardiogram, a startup seeking to make the sensor data from the Apple Watch more meaningful and useful has received $2 million in seed funding to extend functionality of its app.
On average, your Apple Watch measures your heart rate more than 200 times each day. Nevertheless, there is very little in terms of analysis. Cardiogram organizes heart rate data from the Health app to provide you with more detail and historical trends.
The Timeline tab shows your heart rate each day, with a detailed view for workouts and a special interface to track spikes related to stress, diet, or exercise. The Metrics tab shows you how your resting heart rate, activity, and sleep are trending over the last few weeks or months.
“If you own an Apple Watch, its native software already measures your heart rate once every 5 minutes, or continuously if you’re working out,” CEO Brandon Ballinger said in a statement.
“That data does get stored in your Health app, but it’s just raw data, and not easily understandable. What Cardiogram does is organize this info in graphs both real-time short-term graphs, like your heart rate over the course of a stressful meeting, and long-term trend graphs, like resting heart rate over time. Plus, we give you full access to the underlying metrics.”
Since March, Cardiogram has been working with a team of scientists from the University of California to see if this data can be utilized to save you from a stroke. The $2 million was raised in a round led by Andressen Horowitz’s new bio fund, led by Vijay Pande to help fund this research.
Each year, more than 100,000 strokes are caused by an abnormal heart rhythm called atrial fibrillation. In atrial fibrillation, electrical conduction in the heart becomes disorganised. The disease is treatable, but many people don’t feel symptoms, so the trick is detecting it in time.
“In the last six months, we’ve gathered more than 10 billion sensor measurements from more than 100,000 contributors. Of course, raw sensor data is not a biomarker, a biomarker is not a diagnosis, and a diagnosis is not a therapeutic,” Ballinger wrote in a blog post.
“Our first clinical application is detection of atrial fibrillation, the most common abnormal heart rhythm, and early tests with UCSF Cardiology’s Health eHeart Study suggest an accuracy above 90% will be attainable, even with consumer grade heart rate sensors.”
In other news, Cardiogram has also announced a new feature called Habits.
Each Habit packages up a behavior that has been shown to work in the medical literature, and pairs it with an objective biomarker. There are three categories: fitness, sleep, and stress. Each category offers a handful of Habits you can try for 14 days. Each day, you can “check off” your daily progress, monitor whether this Habit is improving key biomarkers.
Like this article? Subscribe to our monthly newsletter and never miss out!