Instagram may be more accurate at diagnosing depression than a doctor, study suggests

A new artificial intelligence program can pick up on the early signs of depression before humans (and even humans who are general practitioners) can — and just by using Instagram. A team of researchers from Harvard and the University of Vermont recently developed a machine learning program that correctly identified which Instagram users were clinically depressed with 70 percent accuracy.In a study published on Aug. 10 from Andrew G. Reece and Christopher M. Danforth, the group applied the machine learning platform to 166 Instagram profiles containing 43,950 photos. The system uses markers to detect if Instagramers are depressed, analyzing aspects like color, metadata, and face detection. Related: Overuse of social media can cause anxiety in teens “Using only photographic details, such as color and brightness,” the study says, “our statistical…


Link to Full Article: Instagram may be more accurate at diagnosing depression than a doctor, study suggests