Study Uses Apple Watch and Neural Network to Detect Hypertension and Sleep Apnea

The Apple Watch has been the subject of many studies which have measured the accuracy of its heart rate scanner and overall just how useful the smartwatch can be in some life-threatening scenarios. Now, a new study done by the University of California San Francisco and a company called Cardiogram indicates that the Apple Watch can detect hypertension and sleep apnea.

The study collected data from 6,115 Apple Watch owners which were then visualised and read by Cardiogram’s deep neural network “DeepHeart.” The neural network was able to correctly identify high blood pressure/hypertension in 82 percent of the cases and sleep apnea with 90 percent accuracy.

It is important to note here that the data from the Apple Watch was completely analysed by the “DeepHeart” neural network. This network was previously used to detect atrial fibrillation, a common abnormal heart rhythm disease. The team behind Cardiogram simply expanded this neural network to multitask and detect hypertension, sleep apnea, and atrial fibrillation. The company first trained the neural network using data from 70 percent of the participants before using it to detect sleep apnea or hypertension in other 30 percent of the remaining participants.

The high accuracy of the results points to a possibility of the Apple Watch being used in the future to detect sleep apnea and hypertension in patients in a cost-effective way on a wide scale.

Cardiogram had previously tested the Apple Watch and its DeepHeart neural network to detect atrial fibrillation in patients with an accuracy of 97 percent. The company plans to use some of its findings in its Apple Watch app for proper diagnosis and treatment of users.