Algorithms can detect patterns in a cardiac MRI that are not visible to the naked eye.

Fatal arrhythmia is one of the most deadly but difficult to predict and therefore prevent.

But what if an AI system could provide some help?

A study published this month in Nature Cardiovascular Research outlines an AI model that has been trained on images of patients' hearts, as well as patients' backgrounds, to have the AI ​​predict whether and when a patient will die of cardiac arrest.

It is currently believed that artificial intelligence is more effective in this task than a doctor, as the artificial intelligence detects patterns in an MRI of the heart that are not visible to the naked eye.

“There are patients who may be at low risk of sudden cardiac death and get pacemakers that they may not need,” said Natalia Traianova, senior author of SciTechDaily and professor of biomedicine at Johns Hopkins University (JOHNS HOPKINS). "High-risk patients do not get the treatment they need and could die prematurely. What our algorithm can do is identify who is at risk of cardiac death and when it will occur, allowing clinicians to decide exactly what to do."

Flatten hidden information

Trajanova and her team used contrast-enhanced images of hearts so that information usually hidden from the naked human eye appears.

Using this technique, the researchers were able to visualize the distribution of scars from hundreds of real patients at Johns Hopkins Hospital, giving them important information about scar patterns that doctors haven't yet had access to.

For his part, the study's first author, Dan Popescu, a former doctoral student at Johns Hopkins University, added that these scars can be distributed in different ways and they say something about a patient's chance of survival, stressing that "there is information hidden in them."

But is this technique accurate?

The algorithms' predictions were further validated in tests with an independent group of patients from 60 health centers across the United States.

The team is now working on developing other systems to detect more diseases.

New AI systems could forever revolutionize how heart disease is diagnosed and treated.