Mood-mapping app helps predict bipolar episodes

Accuracy and typing speed on a smartphone can be used to predict manic and depressive episodes in people with bipolar disorders, researchers reported in a recent article published in Bipolar Disorders.

The researchers developed a unique smartphone app that collects users’ typing movements—or typing kinematics—which indicate changes in an individual’s mood, behavior, and thought.

The researchers anticipate that pinpointing such changes in real time could detect a person’s “vulnerability points,” when intervention is most necessary, particularly during periods of risk in between treatment visits.

The app, BiAffect, is now available for free in the App Store for use in iOS (iPhone) devices. (An Android version is also planned.) The app monitors the user’s “keyboard dynamics metadata,” including typing speed and rhythm, mistakes in texts, and the use of backspace and autocorrect. The metadata of how the user typed, not what they typed, can be analyzed to identify digital biomarkers of manic and depressive episodes of bipolar disorders, researchers said.

“We think that this crowd-sourced, app-based study will soon lead to digital technologies that act as an ‘early alert system’ for people with bipolar disorder to help them see manic and depressive episodes coming, and take action to mitigate the effects of those episodes. Just being aware of them is a step forward for the millions who live with this mood disorder,” said Peter Nelson, PhD, professor of computer science and dean, University of Illinois at Chicago (UIC) College of Engineering, Chicago, IL.

The current purpose of the app is to collect and analyze users’ keyboard metadata. Once the researchers gather enough data, they anticipate that in the next few years, they’ll be able to develop personalized “early warning systems” to detect when an individual user may need more intensive care.

“Unobtrusively monitoring health from an iPhone combines low-cost scalability with far-reaching impact to potentially improve the lives of millions of people,” Dr. Nelson said.

Common behaviors

In an earlier pilot study of 31 participants, researchers showed that altered keystroke dynamics correlated with depressive and manic episodes in people with bipolar disorder. In this study, they reached 90% compliance and logged 27,125 keystrokes.

“During a manic episode, people with bipolar disorder often exhibit common behaviors such as talking really, really fast and acting in an impulsive manner,” said co-lead investigator Alex Leow, MD, PhD, associate professor of psychiatry and bioengineering, UIC College of Medicine. “So it is natural that they also exhibit similar abnormalities in non-verbal communications that are typed on their phones.”

For instance, the spell-check feature prompts smartphone users to pause and consider whether to edit or accept suggestions made by autocorrect—or users can ignore it and simply keep typing.

“People in the midst of a manic episode frequently have diminished self-control, so it is not surprising that our pilot data suggested that some of them tend to blow through the spell-check alerts,” Dr. Leow said.

Similarly, typing a long message may seem laborious for individuals during a depressive episode, so such messages tend to be shorter, she explained.

Users who download BiAffect must first opt into the researchers’ study to identify digital biomarkers of bipolar disorder and to further refine and improve the app. Users will be able to view their own metadata, including their phone usage over time, number of keystrokes, use of spellcheck, and more.

‘Virtual biomarkers’

But will users self-consciously alter their typing dynamics knowing that BiAffect is monitoring them, and potentially throw off the results?

Dr. Leow acknowledged that this is possible, but pointed out that the app collects keyboard metadata passively and unobtrusively. So users may very well type on their smartphones normally, without a constant awareness of the app.

On the other hand, she added, awareness of the app might even help users.

“It was indeed part of our original hypotheses that by providing the right kind of objective information with real-time monitoring, we will be able to help patients form better insight into their own illness,” she told MDLinx.

But the app isn’t just for people with bipolar disorder, Dr. Leow noted.

“We want people without mood disorders to use the app as well so that we can better understand keystroke dynamics in healthy adults versus those with bipolar disorder,” she said. “This will allow us to further hone in on the ‘virtual biomarkers’ of bipolar disorder or even mood in general.”