If you use Instagram a lot, then you may be interested to know that a new algorithm created by researchers from the universities of Vermont and Harvard can analyze your pictures and guess whether you are going through depression or not, without consulting a specialist for to detect this problem.
The researchers used artificial intelligence and feed analysis Instagram of several hundred users to help the system learn how to detect depression based on the pictures published by the users, they affirming that their system is as accurate as possible, at least based on the tests done so far.
The researchers claim that their algorithm it analyzes the color of the pictures, the meta data of its components and the analysis of the user's face, and based on these points of information, the symptoms of depression can be detected in approximately 70% of the analyzed cases, even before the doctors confirm this diagnosis.
The system of researchers a analyzed 44.000 photos of over 166 users to be able to learn how to correctly detect the signs of depression, and although its accuracy is quite high, for now it is not presented as a 100% safe system, but only as one that can raise alarm signals in relation to possible mental problems of users.
Using Instagram data from 166 individuals, we applied machine learning tools to successfully identify markers of depression. Statistical features were computationally extracted from 43,950 participant Instagram photos, using color analysis, metadata components, and algorithmic face detection.
According to a study based on the analysis, the researchers claim that depressed people tend to frequently use blue-gray or black-and-white color filters for the pictures they upload, while healthy people generally use "warm" filters, with tones warm
