The BiAffect app identifies manic and depressive behaviors

Psychologists who found links between the way we type and our moods developed an app to track them. The BiAffect app uses the iOS ResearchKit platform to access an iPhone’s many sensors and usage rates. In essence, the app monitors the way users interact with their smartphone and how they communicate to identify telltale signs of mood swings associated with bipolar disorder.

Bipolar disorder sufferers often go through two distinct phases: manic and the depressive, which are opposite moods with corresponding effects on the way people interact with people and things around them. BiAffect particularly looks at the way people type and talk to their phone. Using iPhone’s TouchForce technology, how hard they press the phone screen can also be a telling sign of the user’s mood.

When an individual is going through a manic episode, self-control is minimized. People in this state usually type longer but less coherent messages. They are also more likely to make mistakes with spelling and grammar. On the other hand, people with some level of depression will type lethargically, are less responsive to messages and more likely to respond with just one or two words. The speed at which people type is also taken into account.

For BiAffect, correlations aren’t always simple. The data the app collects goes through the DeepMood architecture, a deep learning machine that fine tunes results. Keystroke dynamics data go through recurrent neural networks (RNN) algorithms—the same mathematics used in handwriting recognition—to improve the accuracy of the results from the app.

Two psychologists from the University of Illinois at Chicago, Alex Leow and Peter Nelson, developed the BiAffect app. The idea was sparked when one of their sons was diagnosed with bipolar disorder. The results of an internal test with 31 participants strongly supported the app’s hypothesis, and BiAffect won the Mood Challenge for ResearchKit.

By using something as widely adopted as the iPhone, BiAffect’s technology can help users better monitor their moods, easily detecting psychological rhythms that can be dangerous if left unidentified.

BiAffect


Lead Image: Young woman using a smart phone via Shutterstock

Psychologists who found links between the way we type and our moods developed an app to track them. The BiAffect app uses the iOS ResearchKit platform to access an iPhone’s many sensors and usage rates. In essence, the app monitors the way users interact with their smartphone and how they communicate to identify telltale signs of mood swings associated with bipolar disorder.

Bipolar disorder sufferers often go through two distinct phases: manic and the depressive, which are opposite moods with corresponding effects on the way people interact with people and things around them. BiAffect particularly looks at the way people type and talk to their phone. Using iPhone’s TouchForce technology, how hard they press the phone screen can also be a telling sign of the user’s mood.