Bipolar mood detection via smartphone

In July, ApplySci described a Northwestern developed phone app that monitors behavior patterns to detect depression.  Now, Venet Osmani at CREATE-NET has announced a similar phone based concept with a focus on bipolar disorder.   A small study has shown that mood changes can be accurately spotted as they occur,  facilitating earlier treatment and better outcomes.

The manic phase of the disease is often characterized by hyperactivity, which can be measured by an accelerometer, GPS device, speech analysis (for rapid speech) and phone records (for frequent conversations).

Patients in the depressive stage usually demonstrate distinctly different behaviors.

Smartphone activity of 12 bipolar patients was monitored over 12 weeks.  They visited the clinic every three weeks, when a conventional mental state evaluation occurred.

The study found that activity and location data gave a good indication of mood, and accurately predicted mood change 94 percent of the time. When combined with call and speech analysis, accuracy climbed to 97 per cent.  According to Osmani, “amost all changes were detected with almost no false alarms.”

WEARABLE TECH + DIGITAL HEALTH SAN FRANCISCO – APRIL 5, 2016 @ THE MISSION BAY CONFERENCE CENTER

NEUROTECH SAN FRANCISCO – APRIL 6, 2016 @ THE MISSION BAY CONFERENCE CENTER