Artificial Intelligence Can Help Doctors Better Detect Heart Attacks
Scientists have developed an artificial intelligence tool that lets doctors determine whether someone is having a heart attack much faster than current methods.
New research published by healthcare firm Abbott shows that its algorithm could enable hospital accident and emergency departments to more accurately identify and treat patients having a cardiac arrest.
The study, which involved researchers from the U.S., Germany, U.K., Switzerland, Australia and New Zealand and more than 11,000 patients, found that AI could provide doctors a more comprehensive analysis of the probability that a patient was having a heart attack.
Agim Beshiri, a senior medical director at Abbott, said: “AI technology has the capability to consider many variables, characteristics and data points and combine them in seconds into meaningful results.
Today In: Innovation
“Because of today’s advancements in computational power and AI applications, healthcare stands to benefit greatly by this approach where clinicians have to do this with their patients every day.”
Developed by a team of physicians and statisticians at Abbott, the algorithm uses machine learning techniques to enable a more individualized calculation of a person’s heart attack risk.
The technology aims to improve and quicken heart attack diagnosis by analyzing extensive datasets and identifying factors such as age, sex and a person’s specific troponin levels (a cardiac biomarker).
Abbott said the algorithm is designed to help address two barriers that exist today for doctors looking for more individualized information when diagnosing heart attacks.
The first is that international guidelines for using highly sensitive troponin tests don’t always account for personal factors, impacting test results.
And the second is that while these guidelines recommend that doctors carry out troponin testing at fixed times, they don’t consider a person’s age or sex and put patients into a one-size-fits-all situation.
However, Abbott’s algorithm differs from existing approaches as it takes into consideration personal factors and troponin blood test results over time.
Beshiri added: “The World Heart Organization estimates that 17.9 million people die from cardiovascular disease each year, and 85% are due to heart attacks and strokes.
“What’s unique about this algorithm is that it harnessed the power of machine learning to identify what factors are most predictive for determining if someone is having a heart attack or not.
“These factors, such as a person’s age, sex or the dynamics of a troponin blood test, are already being captured when someone enters the hospital with symptoms of a heart attack. The study found that the algorithm helps look at how these variables interact at that moment in time – providing a more individualized and precise calculation.”
Zift Solutions in partner with DELL Technologies
Nicholas Fearn, freelance writer