Your smartphone knows a lot about you, but can it tell when you're feeling depressed? A new app claims to be able to do just that by tracking how you type and offering support if you need it.
The BiAffect app, created by psychiatrists Alex Leow and Peter Nelson from the University of Illinois, is said to detect early warning signs of depression, bipolar disorder, and manic episodes by looking at how hard and fast users hit the keypad as well as whether they take notice of spellcheck.
According to a study of mobile phone meta data, users who were suffering from depression would type slower and craft shorter messages, while those having a manic episode would type faster, use the backspace more often or ignore typos altogether.
"During a manic episode, people with bipolar disorder exhibit some common behaviours, such as talking really, really fast, with diminished self-control and flight of ideas," Leow said.
"People in the midst of a manic episode commonly have reduced impulse control, so it is not surprising that our pilot data supported that they tend to blow through the spell-check alerts."
The app takes advantage of Apple's ResearchKit, an open-source tool for developers to gather data from the iPhone's sensors for medical research, and a tool called DeepMood to analyse their keystrokes using neural network algorithms. The creators hope that by tracking a user's mood it can alert them to early warning signs before it's too late.
However, if you're paranoid about privacy or being spied on, the app creators stress that using this technology doesn't track anything being typed into the phone or what's on screen. It simply detects the movement.
BiAffect was the brainchild of Peter Nelson after his 24-year-old son was diagnosed with bipolar disorder. He wanted to find a way to help his son cope with the brain condition which can cause shifts in mood and energy, as well as the millions around the world who also suffer.
According to Bipolar UK, as many as 5% of the UK population are on the bipolar spectrum and on average it takes over 10 years to receive the correct diagnosis, before this misdiagnosis occurs an average of 3.5 times.
The BiAffect app is planned for release in the iOS App Store later in the year and anyone who is interested can sign up beforehand to receive updates.