New Delhi: Artificial intelligence is better than humans at evaluating cardiac ultrasounds, the leading test of overall heart health, found the most rigorous trial yet conducted on the subject on Wednesday.
While previous research has illustrated the potential power of AI models to read medical scans, the authors of the new US study said it is the first randomized, blinded clinical trial to heart health.
“There is a lot of excitement around AI,” but rigorous testing remains essential, says the study’s lead author. david ouyang he told AFP.
This successful trial “really strengthens the case that we are now ready for prime time,” added the cardiologist at Cedars-Sinai Medical Center in Los Angeles.
Ultrasounds of the heart, also known as echocardiograms, are performed on patients by sonographers, who typically provide an initial evaluation of the scan before turning it over to a cardiologist.
The new study, published in the journal Naturepitted an AI model against sonographers to see who would give the most accurate initial assessment.
Both evaluated the ultrasound for what is called the left ventricular ejection fraction, which measures the ability of the heart to pump blood to the body in the space of one beat.
The test is the main way to measure how well a heart is working. It is used to find out if patients have had a heart attack or if they will be able to undergo serious treatments such as the implantation of a defibrillator.
For the study, nearly 3,500 cardiac ultrasounds were randomly divided between sonographers and the AI model.
Their assessments were then evaluated by cardiologists, who did not know which came from humans and which from the AI model.
– ‘Exciting’: Cardiologists made a substantive change in more than 27 percent of sonographer assessments and nearly 17 percent of those made by the AI model, the study found.
“AI was faster, more accurate and indistinguishable to cardiologists,” Ouyang said.
There is a “tremendous shortage” of sonographers in the United States and around the world, and this would save them valuable time, he added.
The AI model, called EchoNet-Dynamic, was trained on nearly 145,000 echocardiograms and uses what’s called deep learning to process vast amounts of data.
The researchers are currently applying for the method to be approved by the Federal Drug Administrationand I hope to do the same soon in the European Union and elsewhere, Ouyang said.
Patricia PelikaA cardiologist at the Mayo Clinic in the United States who was not involved in the research, told AFP the study was “exciting” and that the integration of artificial intelligence tools will increase efficiency and standardization.
french cardiologist Florian Zores He said the study was done well, but the technology would not be as useful in France, where cardiologists do the initial evaluations of cardiac ultrasounds.
