Image ID 25070396 © Stevanovicigor | Dreamstime.com
Using a novel artificial intelligence-based approach, researchers at Johns Hopkins University have come up with a “foreseeing” model that can predict when a patient could suffer from a heart attack.
The technology, which is built on raw images of diseased hearts and patient backgrounds, could help doctors improve decision-making and increase the chances of survival from sudden or fatal cardiac arrhythmias.
According to the study, sudden cardiac death accounts for up to 20% of all deaths worldwide, with physicians puzzled at how to tell who’s most at risk.
Now, with this AI algorithm, it can be determined who is at risk of heart attacks and when it could possibly occur, allowing doctors to put measures in place to mitigate the risks.
In fact, the technology is so precise, it can predict with high accuracy the chance of a sudden cardiac attack over a decade into the future, and even pinpoint when it’s most likely to occur.
This is because the AI, which looks over hundreds of contrast-enhanced cardiac images from patients, can piece together patterns from cardiac scarring that may not be visible to the naked eye.
“The images carry critical information that doctors haven’t been able to access,” said first author Dan Popescu.
“This scarring can be distributed in different ways and it says something about a patient’s chance for survival. There is information hidden in it,” he explained.
Following that, the team trained a second AI network to sieve information from 10 years’ worth of clinical patient data, using 22 data points including a patient’s age, weight, and prescription drug use.
Together, the system could predict heart attacks significantly more accurately than doctors, and even passed the test when it used in independent patient cohorts from 60 health centers across the country.
Trayanova, co-director of the Alliance for Cardiovascular Diagnostic and Treatment Innovation, said the program could reshape clinical decision-making regarding arrhythmia risk.
“It epitomizes the trend of merging artificial intelligence, engineering, and medicine as the future of healthcare.”
[via
http://www.designtaxi.com/news/418351/This-AI-Can-Predict-If-Someone-Will-Have-A-Heart-Attack-10-Years-Ahead/
Leave a Reply