An AI approach that can accurately predict when will one have a cardiac arrest. Find out more!

A study by researchers at Johns Hopkins University has found a new artificial intelligence (AI) approach to predict whether or not a patient could die of cardiac arrest.

This system is based on image data of patients' diseased hearts and their histories and the potential to change the clinical decision-making and improve the survival from unexpected and lethal cardiac arrhythmias, which is considered to be the deadliest and most confusing illness in medicine.

According to a senior author, Ms Natalia Trayanova, the Murray B.Sachs, Professor of Biomedical Engineering and Medicine, arrhythmia-related sudden cardiac deaths worldwideaccounts for up to 20 per cent death worldwide and people have little awareness about who’s at risk or what’s the prime reason for it to occur.

She said, "There are patients who may be at low risk of sudden cardiac death getting defibrillators that they might not need and then there are high-risk patients that aren't getting the treatment they need and could die in the prime of their life. What our algorithm can do is determine who is at risk for cardiac death and when it will occur, allowing doctors to decide exactly what needs to be done."

These algorithmic predictions rely on cardiac scarring induced by the heart disease which usually leads in deadly arrhythmia.

The deep learning technology is known as Survival Study of Cardiac Arrhythmias Risk (SSCAR) where they have trained the algorithm to identify the patterns and associations not evident to the human eye. For this they'veusedcardiac pictures that reveals about the scar distribution from hundreds of actual heart patients with scarring at the John Hopkins Hospital.

The algorithm can determine who is at risk of heart death and when it will happen, allowing doctors to decide exactly what should be done.

This team is the first one to give a personalised assessment to each heart patient and these factors have accurately predicted the chances of sudden cardiac arrest and even when it will occur in the next ten years. The current clinical image analysis only identifiesbasic features like volume and mass underutilizing what the new study shows is important information.

The research team has also trained a second neural network to learn from 10 years of clinical patients’ data inclusive of their personal details like their age, weight, race and the prescription drug use, and here the predictions were moreaccurate than the doctors were even validated properly suggesting that it canbe adopted anywhere.

A former PhD student at JohnHopkins Hospital, Dan Popescu said, “The images carry critical information that doctors haven’t been able to access. 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 further added.

(With inputs from agencies)