Created by Materia for OpenMind Recommended by Materia
3
Start Dr. Instagram: Will Apps and Social Networks Become the New Doctor’s Offices?
13 September 2017

Dr. Instagram: Will Apps and Social Networks Become the New Doctor’s Offices?

Estimated reading time Time 3 to read

Since their successful irruption in 2010, mobile phone applications have changed our way of interacting and relating with the world. We no longer ask people in order to find an address, nor make a phone call to arrange to meet up with friends, nor do we save photos of our trips in albums. All this we now do with apps. And with the appearance of applications capable of analysing and interpreting content and photographs, our way of relating to medicine has also begun to change. We have started looking for advice, “clinical” guidelines and even diagnoses outside the doctor’s office and within our smartphones.

Researchers from Vermont and Harvard found that people with depression upload more blue, grey, and dark pictures. Credit: PublicDomainPictures

Recently, researchers from Vermont and Harvard have published the results of a study in which they tested an algorithm—software—that processed Instagram photographs to detect clues and warn of a potential depression disorder in the user. As Chris Dantforth, co-director of the University of Vermont Computational Laboratory and lead researcher, explains, “we found that participants with a diagnosis of depression were uploading more blue, grey, and dark pictures with fewer faces than those that the non-depressed subjects were posting. But the most important thing is that these results were achieved before the patients were diagnosed.” And not only that, but the program achieved a success rate of 70% compared to an average diagnosis accuracy of 42% in the medical clinics.

A few months earlier, a team at Stanford University presented another algorithm capable of identifying more than 2,000 types of lesions and dermatological ailments from photographs of the affected skin region. In the tests to which it was subjected, the algorithm matched or surpassed the results obtained by 21 specialists when they evaluated the same images. The user can take a picture of an area “suspected of injury” and know if a visit to the doctor is warranted.

Another of the areas in which this type of app has a large presence is in nutrition. Nowadays, there are numerous applications that analyse and value the nutritional quality of photographs of food that users upload to social networks, or that screen and follow the diet of a person from these digital images and make dietary recommendations.

An early detection instrument

To what extent can these types of apps change the practice of medicine? “I think that this kind of analysis of content uploaded to social networks will become an increasingly common and important tool for physicians,” says Dantforth, “but as an instrument for the early detection of symptoms, not for making diagnoses, at least not in the short term.”

Meanwhile, other experts suggest that this type of medical or clinical application can serve and assist medical practitioners to carry out a more exhaustive and immediate, and therefore more effective, monitoring of each patient. And there are even those who envisage the possibility of a future Orwellian scenario, with the implementation of state programs for the early detection of diseases and disorders through the systematic analysis of content uploaded to social networks by users.

A team at Stanford University presented an algorithm capable of identifying more than 2,000 types of dermatological ailments from photographs. Credit: Niels Olson

What is the point of view of medical professionals? Do they see it as a useful tool or as a dangerous interference? “I don’t believe that today there are any professionals who do not contemplate the possibility of taking advantage of information technology,” says Dr. Diaz Rodríguez, Clinical Head of the Emergency Department of Santa Creu Hospital and Sant Pau of Barcelona; “whether it be for diagnosis-monitoring or even for the training of other physicians; in fact the “clinical simulation” is being postulated as the great tool for the training of health care professionals in the future.”

What’s more, doctors find advantages in these apps. “They have many positive aspects,” argues Dr. Diaz Rodríguez, “for example, telematic monitoring of arrhythmias or of clinical variables such as blood pressure, blood glucose, etc., especially in patients with reduced mobility.”

However, there is still much controversy over the use of apps in this field. There have already been several voices that have warned about the risks and dangers that their use would entail, especially considering the precedent established by the Internet and its free and immediate access to sources, encyclopaedias and medical websites, but also to others of more dubious credibility. Raise your hand if you have ever self-diagnosed after consulting your symptoms on the Internet. In fact, a study carried out in 2015 by the British medical authorities found that almost 25% of Internet users have done just that. “It is an unwanted consequence,” Dantforth asserts. “What should be very clear is that this type of diagnosis is only the first step and that consultation and follow-up by a doctor is always necessary.”

“Doctors are becoming more accustomed to the fact that the patient, when he comes to the appointment, has already consulted information on the Internet, which has its advantages and disadvantages. In many cases false information must be refuted, but on the other hand it requires doctors to be continuously up-to-date so as not to dismiss results and investigations that are real,” says Dr. Díaz Rodríguez. He goes on to add that in the specific case of apps and social networks, the most troublesome aspect, “in addition to avoiding unconfirmed opinions, is how to preserve the patient’s right to privacy.”

Miguel Barral

@migbarral

Comments on this publication

Name cannot be empty
Write a comment here…* (500 words maximum)
This field cannot be empty, Please enter your comment.
*Your comment will be reviewed before being published
Captcha must be solved