Have you ever wondered if a friend is okay when they use the 'inkwell' filter too much on Instagram? New technology may be able tell you.
US researchers have developed software that can tell if someone's depressed based on what they put on Instagram.
It can pick up on early signs of depression before humans - even doctors. Researchers say in tests, it correctly identified users who were clinically depressed with 70 percent accuracy.
The artificial intelligence analyses colours, metadata and face detection to determine signs of depression. Based on earlier research, it knows that darker, greyer colours are more often used by depressed people.
"Using only photographic details, such as colour and brightness," the study says, "our statistical model was able to predict which study participants suffered from depression, and performed better than the rate at which unassisted general practitioners typically perform during in-person patient assessment."
The study says research showed general practitioners correctly diagnosed depression in only 42 percent of cases.
The programme assessed a number of other details to identify depression:
A snapshot of the study showing colour comparisons (Supplied)
The researchers, however, say the findings should not be considered "enduring facts" but as a "methodological foundation upon which to build and refine subsequent models".
One problem the researchers note is that changing trends on social media might eventually render their findings obsolete.