Why your last selfie could predict if you suffer from a heart problem

The word selfie (or selfie) came into our lives in 2014, when Ellen DeGeneres, Brad Pitt, Jennifer Lawrence and other celebrities decided to take a 'selfie' at the Oscar Awards Ceremony, which remains one of the best-known celebrity photos in history. Of course, the selfie (something like the self-portrait) already existed, although it has proliferated during this 21st century thanks to the use of social networks.

Today, the most normal thing in the world is to take the front camera of the mobile, smile and push the button. So much so that in January of this year the so-called 'Selfie Guide' emerged, in which tips are established to take the perfect photo and not die trying, because there really are many accidents during that moment of ego worship. Do you remember the last selfie you took? Did you look handsome? Well, now, in addition to checking if you have wrinkles around your eyes or if you need a shave, you can learn more about your health.

Some facial features (earlobe folds, gray hair …) are related to the risk of heart disease

A recent study published by the European Society of Cardiology has found that we could use selfies for this purpose, reports 'Mel Magazine'. The study, which was published in 'European Heart Journal' discovered that a deep learning computer algorithm could detect coronary artery disease by analyzing four photographs of a person's face. How? As they explored, there are several facial features that tend to be associated with heart disease: some have to do with aging (graying hair or wrinkles), while others are slightly more difficult to determine (earlobe folds or fat deposits under the skin).

In heart disease, gender does matter

EC Brands

They are small traits that doctors often overlook when diagnosing heart problems. In this case, researchers from the Brain and Brain Institute Cognition at the Department of Automation at Tsinghua University in Beijing They developed a computer program that could analyze a person's blood vessels through those four photographs. In one test, the algorithm correctly detected heart disease in 80% of the cases and the absence of this problem in 61% of the rest.

The algorithm correctly detected heart disease in 80% of cases as well as the absence of this problem in 61%

Not perfect or infallible of course but it is an optimistic start. While the particular study was conducted in a laboratory with trained nurses, they reported, it seems feasible that patients could use similar technology in the future. For example, a patient could take multiple selfies and send them to a doctor who, using an improved version of the algorithm, could analyze the photos and determine if the patient needs more tests detection.

The main problem at the moment is related to the ethics of the diagnostic tool and how the use of technology could threaten the privacy of some health data. A problem that, for now, while Artificial Intelligence advances, does not seem easy to solve. But in the meantime, the researchers have more work to be able to improve the algorithm, testing it with a wider group of patients to increase your accuracy. You may not be a big fan of selfies, but who knows, in the future taking a photo may respond more to an attempt to save your life than to a way of capturing time in an egocentric way.