Study co-authors Andrew Reece and Christopher Danforth considered over 43,000 Instagram photos from 166 people, 71 of whom had previously dealt with reported depression.
The study, in EPJ Data Science, looked at nearly 44,000 posts from 166 people (71 of them depressed) using color analysis, metadata, and face detection software.
Researchers created an algorithm that tries to detect depression based on certain qualities in photos posted to Instagram. Compare that to general practitioners' rates for correctly diagnosing depressed patients, which studies have found hover around 42 percent.
Oxford University employee in court over United States murder
Lathem was an associate professor of microbiology and immunology at Northwestern, but was sacked after he was taken into custody. After the hearing, Warren's lawyer, Ariel Boyce-Smith, said: "Mr Warren is agreeable to being transported to Chicago".
Moreover, the study also led researchers to believe that fewer faces per photo were an indication of a mental illness like depression. "Depressed people also tended to prefer Instagram's Inkwell filter, which turns a color image into black-and-white, whereas healthy participants preferred the Valencia filter, which gives photos a warmer, brighter tone". With depression on the rise among US teens and young adults (also Instagram's core demographic), better methods for diagnosing mental health problems are more important than ever.