Cardiogram launches new Apple Watch features with a view toward heart health

By Jeff Lagasse
Share

Cardiogram, a startup working on algorithms to make the Apple Watch’s heart rate data clinically actionable, is launching two new features for the Apple Watch that aim to help people stay more active this summer: Leaderboards and Workout Zones.

Leaderboards is geared toward those who want to inject some friendly competition into their active lifestyles. Friends can compete with one another and track each others’ progress throughout the day, in turn challenging the other to be more active -- helping to increase exercise levels, and potentially lead to healthier lives.

The feature also lets users compare their exercise levels to other people in their age range. This effectively creates a goal for the user -- to match or surpass the average exercise level of their age group, or at the very least make some progress toward it.

Workout Zones, meanwhile, allow users to track the intensity of their workouts. For each workout, a person can now see how long they spent in each of their target heart rate zones. The American Heart Association recommends at least 30 minutes of moderate-intensity aerobic activity, which translates into roughly 50-70 percent of max heart rate at least five days per week for a total of 150 minutes -- or at least 25 minutes of vigorous aerobic activity, at 70-85 percent of max heart rate, at least three days a week for a total of 75 minutes.

Alone, Apple Watch is projected to create more than 2 trillion heart rate measurements this year. But that’s just raw data. Cardiogram interprets what the heart rate means: Whether a spike in heart rate corresponds to what a person is eating, for example, or whether it’s due to stress of an abnormal heart condition.

To take things a step further, Cardiogram is also running an N=14,011 study with UCSF Cardiology, using the Apple Watch to predict and prevent heart disease. In May, the company presented a clinical study showing that its deep neural network, DeepHeart, can detect atrial fibrillation with 97 percent accuracy compared to an in-hospital gold standard, using Apple Watch’s heart rate sensor alone. Similar to how Siri or Google speech recognition transforms a raw audio signal into a sequence of phonemes, DeepHeart transforms raw heart rate signal into a sequence of health risk scores.

Cardiogram plans to invest in more medical research, and expand to other conditions associated with heart rate variability.