It seems there’s not much AI can’t do these days. Whether it’s drive a car, improve music discovery, retouch photos or narrate the world to blind people. And now machine learning can even look through your Instagram feed to decide whether you are clinically depressed …
A research study published today examined the Instagram feeds of 166 individuals, looking for so-called markers of clinical depression. The study’s authors claim that the evaluation performed by the AI system was even more successful than general practitioners in diagnosing the condition.
Statistical features were computationally extracted from 43,950 participant Instagram photos, using color analysis, metadata components, and algorithmic face detection. Resulting models outperformed general practitioners’ average unassisted diagnostic success rate for depression. These results held even when the analysis was restricted to posts made before depressed individuals were first diagnosed.
The AI system examined a number of factors ranging from the colors found in photos (including whether or not filters were applied by the user) to the frequency of posting.
In studies associating mood, color, and mental health, healthy individuals identified darker, grayer colors with negative mood, and generally preferred brighter, more vivid colors [16, 17, 18, 19]. By contrast, depressed individuals were found to prefer darker, grayer colors […]
Depression is strongly associated with reduced social activity [20, 21]. As Instagram is used to share personal experiences, it is reasonable to infer that posted photos with people in them may capture aspects of a user’s social life. On this premise, we used a face detection algorithm to analyze Instagram posts for the presence and number of human faces in each photograph. We also counted the number of comments and likes each post received as measures of community engagement, and used posting frequency as a metric for user engagement.
Interestingly, engagement measures found that both fewer likes and more comments were indicative of a depressed person.
Apple is heavily involved in AI work, with Siri the first obvious expression of this work. It had been thought that its extreme secrecy was hampering its efforts to recruit the best AI researchers, but a policy reversal meant that Apple AI teams were able to publish from the end of last year. The first published paper was on image learning.
Apple finally joined Instagram just this week.