Stress is a growing mental health problem that can have severe consequences for human health. The development of stress prediction tools would benefit public health greatly by enabling policy initiatives and early stress-reduction interventions. With the advent of mobile health technologies such as smartphones and smartwatches, it is now possible to collect objective, real-time, and continuous health data.
Researchers from the University of Waterloo in Canada launched a study to pilot the collection of heart rate variability data from the Apple Watch ECG sensor and develop a stress prediction tool. Using the Apple Watch Series 6’s ECG sensor, the researchers discovered a close relationship between ECG data and heart acceleration and deceleration capacity.
Based on ECG readings and data from 33 study participants’ stress questionnaires, they used Random Forest (RF) and Support Vector Machines (SVM) to model stress. Machine learning algorithms were then developed using this information to create a prediction model.
The prediction model
To further investigate the prediction model, the researchers divided the data they collected into various socio-demographic groups. Overall, the RF model outperformed the SVM model, with results within the low end of the state-of-the-art. According to reports, the stress models have higher precision but lower recall.
“Our models showed specificity in their capacity to assess “no stress” states but were less successful at capturing “stress” states.”
According to the study, the Apple Watch has a “promising” potential for stress prediction. Since the gadget also gathers additional health data, such as sleep and activity data, one could add more data points to stress models to improve their predictive accuracy.
Overall, the findings suggest the possibility of creating a stress prediction tool using the ECG sensor data from an Apple Watch with additional development and improvement. A wearable device that can monitor stress in real time would allow people to respond to changes in their mental health sooner.
The researchers believe that the Apple Watch could help mental health care by providing activities such as breathing exercises to counteract stress signals and responding to changes in mental health early.
Large-scale data gathering from such devices would also assist in informing public health initiatives and regulations. Competing devices such as Samsung, Fitbit, and Garmin have long included a stress score feature, but Apple has yet to include one in its Health app.