Depression has the power to affect most areas of a person's life, even smartphone use, and researchers in the US have come up with an app that's able to recognise when someone's at risk. The underlying hypothesis is that the lower our mood, the more likely we are to spend time moping around on our phones.

The app, developed by academics at Northwestern University, is called Purple Robot, and it uses a number of different indicators to gauge a person's mood, including app usage and location logging. To test the device, 40 users were monitored over the course of two weeks, before being asked to take the standardised PHQ-9 depression test. The app identified 87 percent of the at-risk individuals flagged by the traditional questionnaire.

Movement - measured via the phone's integrated GPS - was a major factor, with those more susceptible to depression travelling around much less. Phone usage was less accurate as an indicator, but was still able to spot 74 percent of the people at risk according to PHQ-9. Now the team wants to see a broader study carried out to investigate whether combining GPS data and usage logs can be as effective as traditional surveys in spotting the signs of depression in patients.

PHQ-9 is not in itself a definitively accurate method of detecting depression, but the results from Purple Robot are enough to suggest there's some correlation there. Future versions of Purple Robot will be able to tap into more smartphone sensors to gather more data.

The team behind the study say apps like Purple Robot could one day be used as "behavioural intervention technologies" to encourage users into habits that will help them avoid depression. "This can improve the identification of depression and the ability of health care settings to allocate resources to those in need and overcome the individual and systemic barriers to conventional psychological treatment," concludes the report, published in The Journal of Medical Internet Research.

One of the advantages of a system like this is that it requires very little input from the user, with everything collected passively - there are no doctor appointments to attend, no surveys to fill in, and no reports to write. In theory, a depression-spotting app could catch the warning signs of someone withdrawing from society and normal life.

"Phones fit into the fabric of people's lives," write the authors. "People tend to keep phones with them all or most of the time, and phones can provide data unobtrusively and with no effort on the part of the user. This capacity offers new opportunities to identify human behaviour patterns associated with depression or other health and mental health disorders."