How Machine Learning Helps Heart Attacks

November 11, 2019 0 Comments

There is a new artificial intelligence (AI) tool that helps doctors determine whether a patient is having a heart attack. Find out more about the research and implications. 

Machine Learning for Heart Attacks

The Research

Implications

The Research

New research published by the international healthcare company Abbott shows that the algorithm they used could allow hospital emergency departments to identify patients having a cardiac arrest more accurately

Researchers from the US, Germany, UK, Switzerland, Australia and New Zealand and more than 11,000 patients participated in the study. The test was to determine whether artificial intelligence technology could provide faster and more accurate heart attack diagnosis.

The algorithm in the AI tool analyses extensive data sets and identifies the variables that predict a cardiac event. The variables include age, sex, a patient’s specific troponin levels and blood sample timing.

In such a way, the AI tool has also enabled a more individualised calculation of a person’s heart attack risk.

Artificial intelligence cpu unit

The research showed that the AI tool could provide doctors with a more comprehensive analysis of the probability that a patient is having a heart attack, faster than the current methods.

Abbott said the algorithm is designed to address two barriers that exist today for doctors when diagnosing heart attacks in individual cases:

  • Firstly, the international guidelines for using sensitive troponin tests do not always account for personal factors, which were proven to have an impact on the test results.
  • Secondly, while these guidelines recommend doctors to carry out troponin testing at fixed times over a period of up to 12 hours, they do not consider a person’s age or sex and it puts patients into one algorithm, leading to no personalised factors being taken into account.

Researchers found when this information is combined through computation, this algorithm has the potential to give doctors more confidence in the results. It gives them a more comprehensive assessment of the probability that a patient was having a heart attack, especially with those patients who presented within the first three hours of their symptoms. 

GP cardiology instruments signs

Implications

The research found that the algorithm performed better than the European Society of Cardiology pathway.

The European Society of Cardiology rule-out pathway includes a rapid assessment algorithm based on high-sensitivity cardiac troponin and sampling guidelines at different timed intervals to rule out heart attacks.

The algorithm developed by the Abbott team is more versatile than existing algorithms, it neither depends on fixed cardiac troponin thresholds nor require testing to be performed at specific time points.

The study concluded that the algorithm’s individualised assessment of the likelihood for Myocardial Infarction could be used as a tool for identifying both low-risk and high-risk patients, especially to benefit from earlier clinical decisions.

Doctor stethoscope and heart button on a desk

According to Abbott, the algorithm was used for research only and is not yet commercially available. The company has also applied for an international patent for the use of the algorithm.

A senior medical director at Abbott mentioned that the AI technology is able to consider many variables, characteristics and data points and it can combine them into meaningful results in a matter of seconds. Thanks to the advancements in computation and AI applications, Australia’s healthcare can benefit greatly with this approach.

Questions?

What are your thoughts on the study?  Please contact one of our friendly team of consultants Today and let us tell you what Locum GP and Permanent GP jobs are available around Australia and whether they could be exactly what you are looking for.

This week’s Gorilla Jobs Update also includes information for our Pharmacy and Medical Imaging divisions. For more useful information please visit our Blogs page.

For more Doctor Jobs, Pharmacist Jobs, Sonographer Jobs, Radiologist Jobs and Radiographer Jobs

Please visit www.gorillajobs.com.au