Selfies Can Help Detect Heart Conditions, Study Finds

Category: Health


Unlocking Word Meanings

Read the following words/expressions found in today’s article.

  1. crease / kris / (n) – a fold on the surface of something

    He smoothed out the creases on his shirt before facing his boss.

  2. red flag / rɛdflæg / (n) – a warning sign that there is a problem or issue with something or someone

    Constant forgetfulness and difficulty concentrating may be red flags for an underlying brain disease.

  3. socioeconomic status / ˈsoʊ si oʊˌɛk əˈnɒm ɪk ˈstæt əs / (n) – a person or group’s social status and economic standing in a society

    A family’s socioeconomic status is usually determined by their total annual income.

  4. fine-tune / ˈfaɪnˈtun / (v) – to make small changes to improve the way something works

    The research team got more accurate results after they fine-tuned their testing tools.

  5. ethnicity / ɛθˈnɪs ɪ ti / (n) – a group of people who come from the same region and have the same language, culture, etc.

    My friend and I may have come from the same country, but we grew up in different regions and are different ethnicities.


Read the text below.

A new study revealed that sending selfies to doctors can help in the early diagnosis of heart disease.

The European Society of Cardiology proposed a method of detecting heart disease by examining photographs of a person’s face. Through this method, selfies sent to doctors can be analyzed with AI to quickly diagnose heart disease and determine its stage. The researchers claim that the new method may diagnose heart disease more effectively than current methods.

The method is based on the fact that some observable changes on a patient’s face, such as earlobe creases and skin wrinkling, can be red flags for heart disease.

To develop the AI, the researchers observed 5,796 patients from eight hospitals between July 2017 and March 2019. The patients’ blood vessels were examined through imaging procedures. Then, radiologists reviewed these images to check for narrowed blood vessels and assess the patients’ risk of heart disease. The patients also had photos of their faces taken by trained nurses, and they were interviewed regarding their socioeconomic status, lifestyle, and medical history. This information was used for the training of the AI.

Afterward, the researchers did additional tests on the AI with another 1,013 patients from nine other hospitals.

According to the researchers, their goal is to create an affordable way for people at high risk of heart disease to assess themselves before seeing a doctor. They added that the AI they developed may be used to make a mobile app in the future.

However, the team said that they still need to fine-tune the method to address its limitations. Before it can be widely used, the AI still needs to be tested on other ethnicities. Experts who have reviewed the study agree, saying that the new method must be validated and improved for larger populations. They also pointed out that using the patients’ photos could raise some concerns about their privacy.

Viewpoint Discussion

Enjoy a discussion with your tutor.

Discussion A

• In your opinion, what factors may cause this method to be inaccurate or unreliable (e.g. technical problems, quality of images)? Discuss.
• Do you think patients should be allowed to perform initial screening for diseases on themselves? Why or why not?

Discussion B

• If you were a patient, would you be comfortable with getting an AI diagnosis? Why or why not?
• If a hospital or clinic uses AI for diagnosis, what can it do to assure patients that the process is safe (e.g. delete photos as soon as possible)? Discuss.