Researchers, using machine learning models, have discovered that fitness trackers, like those on Fitbit, can help detect episodes of mania and depression in people with bipolar disorder.
Researchers from a hospital in Boston have found that leveraging data from sensors in digital devices, such as smartwatches, can accurately detect mood swings in people with bipolar disorder (BD).
See related: FDA: Approves sleep apnea detection on Apple Watch
Bipolar disorder is characterized by severe mood swings, ranging from a person's depression to mania. Depression is often described as a feeling of intense, persistent sadness, hopelessness, or apathy. Mania, on the other hand, can include intense emotions and psychotic symptoms.
The study involved 54 adults with a diagnosis of BD who wore Fitbits for nine months. The researchers applied machine learning models to the data collected over two-week periods to identify changes in mood. The findings showed that the Fitbits could accurately detect mood changes, including episodes of (hypo)mania and depression.
Using a machine learning algorithm, the researchers were able to identify symptoms of depression with over 80% accuracy and (hypo)mania with over 89% accuracy. This approach is particularly useful, as algorithms could predict when BD patients are likely to experience a depressive or manic episode between visits to psychologists or doctors.
Read more: New details about Fitbit's first smartwatch leaked
How do smartwatches track mood changes?
Researchers are analyzing passive sensor data collected by fitness trackers to detect when a person is experiencing depression or manic episodes. This data is collected without the user's direct intervention and includes information such as activity, heart rate, sleep patterns and blood oxygen levels, as recorded by devices like Fitbit or Apple Watch.
Depending on the type of episode, whether depressive or manic, certain physical changes are observed. For example, those experiencing a depressive episode may move less or sleep more than usual. These types of data are passively collected by smartwatches and fitness trackers.
See also: Fitbit: Will remove Pandora and Deezer support from older smartwatches
In contrast, those experiencing a manic episode may experience an increased heart rate, more activity, and less sleep. However, symptoms vary from person to person.
Source: cybernews
