Although medicine is seriously developed and continues to develop today, it is still not possible to prevent some things. At the top of these things is sudden cardiac arrest. Studies show that one out of every 5 people who lost their lives today dies due to sudden cardiac arrest.
Many people are unaware of whether they are at risk of sudden cardiac arrest. Here, scientists have developed a new algorithm that will allow people to learn whether they are at risk of sudden cardiac arrest and take precautions accordingly. The algorithm developed by scientists from John Hopkins University makes use of deep learning technology.
can predict whether there is risk and when exactly the heart may stop within 10 years:
‘Survival Study of Cardiac Arrhythmia Risk (Cardiac Arrhythmia Survival Study) – SSCAR) uses neural networks that create a personalized assessment of survival for each patient with heart disease. This algorithm, which is a first in the world, calculates the probability of sudden cardiac arrest in 10 years with risk measurements and presents the most likely time with ‘high accuracy’.
The algorithm used images of the actual damaged hearts at the University hospital where the damage distribution can be seen to train the algorithm. The images presented to the algorithm with high contrast paved the way for the algorithm to see details that cannot be seen with the naked eye. The algorithm makes its predictions by comparing the images it has stored in memory with the images presented to it.
The algorithm’s predictions were able to produce more accurate results than the doctors’ predictions in each test performed. In addition, tests with different cardiac histories and different imaging data of individual patient groups from 60 health centers in the USA also gave high accuracy results. The new algorithm has become one of the most important steps in the role of artificial intelligence in the health sector.
27357
