APR 11, 2022 12:00 PM PDT

AI Accurately Predicts if Someone Will Die of Cardiac Arrest

WRITTEN BY: Ryan Vingum

Cardiac arrest occurs when the heart suddenly stops working, usually leading to the inability to breathe or a loss of consciousness as the result of reduced blood flow to the brain. If not treated immediately, cardiac arrest can lead to serious disability or death.

There are several factors that can put someone at increased risk of experiencing sudden cardiac arrest. The three most common factors however, include the following:

  • Arrhythmias. Characterized by abnormal heartbeats, arrhythmias are often caused by changes in the electrical signals the brain sends to the heart, causing the heart to beat out of rhythm. 
  • Cardiomyopathy. This condition essentially refers to an enlarged heart, where heart muscle becomes thick. Thicker muscle prevents the heart from pumping effectively. 
  • Coronary artery disease (CAD). This is the most common type of heart disease, and is characterized by a buildup of plaque blockages in the heart. 

Despite common risk factors being known, determining who is most at risk of cardiac arrest and when it might happen is still challenging for clinicians. Researchers at Johns Hopkins University are looking to leverage artificial technology to support clinical decision making by trying to predict patients most at risk of dying from a sudden cardiac arrest and potentially improve the odds of survival. Their efforts are published in a recent edition of Nature Cardiovascular Research.

The research team points to discrepancies in how care is provided and the challenges in giving people the care they need for cardiac arrest. For example, some people who are at high risk of sudden cardiac arrest simply don’t get the care they need. 

To address this problem, researchers developed an artificial learning technology called the Survival Study of Cardiac Arrhythmia Risk (SSCAR) to build profiles and risk assessments for individual patients with heart disease. These assessments include highly accurate predictions on when someone is most likely, if at all, to experience cardiac arrest. What’s more, the predictions were far more accurate than a physicians following clinical validation of patients from over 60 health centers in the U.S. 

To train the AI, researchers used enhanced heart images that include scarring from actual cardiac patients at Johns Hopkins. The AI is able to then detect and trace patterns in scarring, for example, that a doctor might not be able to notice. This ability is what lends to more accurate predictions about cardiac events.  

Sources: Science Daily; Nature Cardiovascular Research

About the Author
Professional Writing
Science writer and editor, with a focus on simplifying complex information about health, medicine, technology, and clinical drug development for a general audience.
You May Also Like
JUL 11, 2022
Earth & The Environment
Volcanic Eruptions Can Help Us Better Understand the Climate
JUL 11, 2022
Volcanic Eruptions Can Help Us Better Understand the Climate
An international team of researchers led by the University of Bath studied the atmospheric response to the January erupt ...
JUL 22, 2022
Earth & The Environment
Scientists conduct first genetic diversity census of common dolphins
JUL 22, 2022
Scientists conduct first genetic diversity census of common dolphins
A recent and comprehensive study out of Flinders University in Australia that was published in BMC Ecology and Evolution ...
JUL 25, 2022
Technology
Tools of the Trade: Ancient tattoos brought to life
JUL 25, 2022
Tools of the Trade: Ancient tattoos brought to life
With little to no information in the archaeological record regarding tattoos, placing their origin has proven difficult. ...
AUG 12, 2022
Earth & The Environment
Humans, not climate, contributed to mammal community similarities 10,000+ years ago
AUG 12, 2022
Humans, not climate, contributed to mammal community similarities 10,000+ years ago
In a recent study published in Nature Communications, an international team of researchers discuss how the increasing ho ...
AUG 05, 2022
Space & Astronomy
First Detection of a Compact, Massive Triple Star System
AUG 05, 2022
First Detection of a Compact, Massive Triple Star System
Two researchers at the Niels Bohr Institute at the University of Copenhagen are studying an usually compact system of th ...
AUG 14, 2022
Space & Astronomy
Looking Back in Space: NASA's Pioneer Program
AUG 14, 2022
Looking Back in Space: NASA's Pioneer Program
This series will explore historic space missions from the start of the Space Age to the present day, including both crew ...
Loading Comments...