How an algorithm learned to identify depression from your photos

Researchers have trained a machine to spot depression from pictures you post on social media, namely Instagram. It seems that intuitively, we tend to link darker, grayer colors with negative moods and brighter, lighter colors with positive ones. How people suffering from depression prefer darker colors is reported in the MIT Technology Review. Clearly this raises the possibility that it could be possible to diagnose depression within the population by analyzing the photos people post to social-media sites. But how reliable could such an approach ever be? The analysis also identified some curious observations. Photographs people rated as happy or sad were only weakly correlated with depression. Depressed individuals were also more likely to post photos with faces, but these photos tended to have fewer faces per photo. 

The finding raises the question as to how well the algorithm might actually identify depressed individuals. The researchers tested it on images posted by 100 individuals. The algorithm correctly identified 70 percent of those who were depressed. That’s significantly better than GPs manage when asked to identify depressed individuals. The work raises the poential for more accurate and prompt identification of mental illness and more effective intervention.