Health News

Instagram Filters Can Show Depression, New Machine Computes

Photo from Wikipedia

Social media is indeed a multi-purpose tool, as it now serves one more important function: diagnosing depression and other mental illnesses.

Andrew Reece of Harvard University and Chris Danforth of the University of Vermont have come up with a machine that uses photo application Instagram to scan for depression and similar conditions, Tech Times reports.

Reece and Danforth’s study found a link between a person’s use of color and his or her mental state.

They theorized that Instagram users who post grayer and darker-colored images are more likely to be or become depressed compared to those who post brighter-colored photos.

The pair then developed a machine that identifies if a person is depressed by scanning his or her Instagram photos.

One photo can have different interpretations. Images on the app can be edited with filters that turn colored photos black-and-white. Other studies suggest that people who are depressed tend to use these filters more often than individuals who are not.

The researchers studied around 170 employees from Amazon’s Mechanical Turk service, all of whom had Instagram accounts. All participants were asked to complete a survey, which included a standard clinical depression questionnaire. They were also requested to share photos from their Instagram accounts.

Out of all the photos, the researchers took about 100 from each subject and asked people to rate them on a scale of 0 to 5, based on how sad, happy or interesting the images looked. The photos were categorized according to saturation, hue and the number of faces in each picture.

These photos went through a machine-learning algorithm that spotted the correlations between image characteristics and depression.

The algorithm discovered that decreased saturation and brightness, along with increased hue, predicted depression. The study also showed that depressed people were less likely to use filters compared to their filter-happy counterparts.

The machine was able to identify depression with a 70% success rate. It is not possible for any method to spot depression with an incredibly high rate of achievement; even clinical surveys and medical experts are not 100% accurate.

The study says, “More generally, these findings support the notion that major changes in individual psychology are transmitted in social media use, and can be identified via computational methods.”

Danforth and Reece say that this technique can be helpful in further understanding mental illness in people, and detecting depression early on for better diagnosis and treatment.

Click to comment
To Top

Hi - We Would Love To Keep In Touch

If you liked this article then please consider joing our mailing list to receive the latest news, updates and opportunities from our team.

We don't want an impostor using your email address so please look for an email from us and click the link to confirm your email address.