Unlike physical ailments, mental health concerns like anxiety and depression are more difficult to notice because they are happening internally, many times without any obvious changes that would be glaring to colleagues, friends, loved ones, or even the individual themselves.
Being able to detect if someone is becoming depressed early on can help that person make informed decisions about their mental health. Now, new research suggests that wearable fitness trackers like the Fitbit can make this a reality by warning wearers of otherwise undetectable physical changes that indicate that depression may be creeping upon them.
The study, which was conducted by a team from Singapore’s Nanyang Technological University, involved a total of 290 adults (average age of 33), who were instructed to wear a Fitbit Charge 2 activity-tracking device for 14 consecutive days. The participants had to wear the device at all times, except for when bathing or recharging its battery.
At the beginning and end of the two-week stretch, participants completed a questionnaire that is widely used to identify people who are becoming depressed. The results of the questionnaires were combined with the data gathered by the wearable devices and these results were used to train a machine-learning-based computer program called the Ycogni model.
When the program was later used to analyze data exclusively gathered by the Fitbits, it proved to be around 80 percent accurate at predicting which individuals were most likely and least to develop depression.
The program determined that participants with more varied heart rates between 2:00 am and 4:00 am, and then again between 4:00 am and 6:00 am were more at risk of developing depression, which falls in line with findings from earlier studies that suggest that variations in heart rate while sleeping could be a valid physiological indicator of depression.
According to the Fitbits, at-risk test subjects also tended to have a wider variation in waking times and bedtimes, which again aligns with other studies that have observed that people suffering from depression aren’t as strict with their sleeping and waking routines.
“Our study successfully showed that we could harness sensor data from wearables to aid in detecting the risk of developing depression in individuals,” said Professor Josip Car, who led the study with associated Professor Georgios Christopoulos. “By tapping on our machine learning program, as well as the increasing popularity of wearable devices, it could one day be used for timely and unobtrusive depression screening.”
Source study: JMIR Mhealth Uhealth – Digital biomarkers for depressions screening with wearable devices: Cross-sectional study with machine learning modeling