A recent study conducted by Andrew G Reece from Harvard University and Christopher M Danforth from the University of Vermont has suggested an innovative idea that can help doctors better diagnose mental health problems by simply viewing their patient’s social media photos like those on Instagram.
Details About The Study Conducted
Over 43,000 pictures from 166 people from Instagram were used in the study on the correlation between their mental health and the evaluation of their social media photo choices. The researchers had also asked the participants about their mental health history with a little below half of the participants, in the past three years, have been found to have been diagnosed with depression.
An algorithm was developed by the researchers to help analyze the various components of the photos collected to determine the subjects’ mental health. This algorithm analyzed the Instagram photos to assess the following:
- The number of people found in the photo.
- Colors, both natural and filtered, in the photo.
- The number of likes and comments on the photo.
Groundbreaking Results Of The Study
It was found that those subjects with depression had fewer people in their photos as well as darker colors and were not as likely to use filters before uploading. However, if they did choose a filter to edit their photos, researchers found that they “disproportionately favored the ‘Inkwell’ filter, which converts color photographs to black-and-white images,” with the other subjects who didn’t have depression often opting for more vibrant filters like the ‘Valenica.’
The difference between a photo uploaded by a healthy participant (left) vs. one suffering from depression (right)
After observing the various photo selections as well as the various factor chosen by the participants like the color and filtering, the algorithm was created by the researchers. This algorithm was used on the photos of the 166 participants to assess their mental health state with an accuracy of over 70% for depression. This breakthrough computer program had also found signs of depression in older photos of participants from before the time they were diagnosed.
More Research Needed For Concrete Proof
However, this should not have you stalking your family members’ Instagram photos for signs of mental health problems just yet. Because while this study did help create an algorithm with an accuracy of over 70%, there are other factors to be considered:
- This is a ‘proof of concept’ study to assess the real-world application.
- The sample size was small with 166 participants.
- The participants were volunteers who were relatively active on Instagram and were willing to submit their mental health history.
According to the author of this study, Chris Danforth who is also the co-director at the University of Vermont’s Computational Story Lab, it’s still unclear as to whether the same outcome can be achieved when applied to average Instagram users. The researchers hope to help encourage the scientific community to conduct more research into the correlation between depression signs and technology for better detection of mental health issues.
In an interview, Danforth said “It shows some promise to the idea that you might be able to build a tool like this to get individuals help sooner,” and can especially be helpful for doctors to diagnose better their patients who come in periodically over months and years. The research conducted is also shown to be observed by doctors in the past as people who suffer from depression often withdraw from social groups and have fewer chances of taking photos with others. Their view of the world can often be dark which may cause them to prefer a darker filter or colors in their Instagram photos.
Future Research On Technology And Mental Health
Surveys have been conducted to find that almost 300 million people around the world suffer from depression and technology such as the algorithm can help individuals the help that is needed. This is even more so to those who are unaware or are reluctant about stepping forward.
Speaking of the future, Danforth said “The end goal of this would be creating something that monitors a person’s voice, how they’re moving around and what their social network looks like ― all the stuff we already reveal to our phones. Then that could give doctors a ping to check in or at least some insight. Because maybe there’s something going on that even the individual doesn’t recognize about their behavior.”
A 2013 study conducted by researchers at the University of Vermont used geotagged tweets to formulate a comprehensive report on the happiest and least happy states using social media as their data source. Similar innovations have started to be used by scientists to understand human nature better and are a great first step towards integrating social media technology and health research for a better tomorrow for us all.