There may be a new way for researchers to get a deeper look into the mental state of patients, social media. While this may sound obvious, a new algorithm can determine if a patient is depressed based upon the images said person posts to Instagram. Images with darker and more negative moods have been linked to a more depressed mental state, which is raising some interesting opportunities in the diagnosis of clinical depression. Doctors may be able to select at risk or depressed patients through simply having a computer sort through various Instagram feeds. This could help stop suicides and allow intervention before drastic choices are made.
The first question that needs to be answered in this technique is just how reliable can a computer algorithm predict the mental state of Instagram Users. Researchers from Harvard University and the University of Vermont have developed this algorithm from significant breakthroughs made in the correlation of image substance to mental state.
“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. Resulting models outperformed general practitioners’ average diagnostic success rate for depression.” ~ Cornell University
[Image Source: MIT Technology Review]
A team of 500 workers from Amazon’s Mechanical Turk service was used to analyze Instagram accounts through questionnaires and surveys, according to MIT Tech Review. From the study of the surveys in correlation to the user’s Instagram accounts, a strong correlation was made between certain photographic aspects in posts. After some pairing down, 170 workers were used in the final study and 70 were found to be depressed based on standard diagnosing practices. Then researchers
After some pairing down, 170 workers were used in the final study and 70 were found to be depressed based on standard diagnosing practices. Then researchers analyzed Instagram posts of depressed individuals to find commonalities. There were many aspects of a photo used in the study, from the obvious hue and subject matter, researchers analyzed the number of faces in photographs to extrapolate a users’ social activity state. In the end, photos with bluer, grayer, and darker tones as well as receiving fewer likes than those posted by healthy individuals were all found to be key indications of a user’s depression.