BiAffect uses neural network algorithms to predict and track mental health
App is designed just for an iPhone and uses Apple’s ResearchKit
Tracks typing speed and how hard keys are pressed when using the keyboard
Also monitors the frequency of the use of backspace and spellcheck
Some 5.7 million US adults suffer from bipolar disorder, however, the only way to treat them during an episode is to watch for signs – but sometimes it is too late.
Now, researchers have developed an iPhone app that uses Apple’s ResearchKit, allowing the technology to track and predict the user’s mood and episodes by analyzing their keystrokes.
Called BiAffect, the technology screens typing speed, how hard keys are pressed and the frequency of the use of backspace and spellcheck, all of which can reveal if the person is experiencing depression or mania.
HOW DOES IT WORK?
BiAffect is specifically designed for iOS and uses Apple’s ResearchKit.
The app is capable of tracking and predicting the user’s mood and episodes by analyzing their keystrokes.
It uses DeepMood architecture that allows the app to analyze keystroke dynamics data in order to infer the user’s mood state using state of the art recurrent neural network (RNN) algorithms.
It monitors typing speed, how hard keys are pressed and the frequency of the use of backspace and spellcheck
The mental health app was developed by a team of psychiatrists, Alex Leow and Peter Nelson, from the University of Illinois at Chicago.
The idea was sparked after Nelson’s 24-year-old son was diagnosed with bipolar disorder as a freshman in college.
‘I began working on this idea many years ago as a way to help my son, and to see it come to fruition with this kind of recognition, and to know that the app will be out there to help people get a better understanding of this disorder is thrilling,’ Nelson said.
Bipolar disorder, which causes extreme mood swings between the emotional highs of manic episodes and low periods of depression, affects approximately 5.7 million, or 2.6 percent, of adult Americans, according to the National Institute of Mental Health.
And diagnosis relies on careful history-taking and examination.
‘The vision for BiAffect is for it to serve as a kind of ‘fitness tracker’ for the brain,’ Leow said.
‘The Mood Challenge helped us to realize this vision, and the finished app will be a first-of-its kind tool for researchers to study mood disorders and even cognitive disorders such as Parkinson’s and Alzheimer’s disease.’
BiAffect uses a DeepMood architecture that allows the app to analyze keystroke dynamics data in order to infer the user’s mood state using state of the art recurrent neural network (RNN) algorithms.
The team, in collaboration with Kelly Ryan, clinical assistant professor of psychiatry at the University of Michigan, had recently completed a pilot study of 30 participants that showed altered keystroke dynamics correlated with depressive and manic episodes in people with bipolar disorder