Scholars of the Cedars-Sinei Medical center have developed a algorithm which will make it easier to identify individuals who are about to be affected by a heart attack.
The new technology accurately identified which patients would suffer a heart attack in five years based on the amount and composition of plaque in the arteries that supply blood to the heart.
The buildup of plaque can cause the arteries to narrow, which prevents blood from reaching the heart, increasing the likelihood of a heart attack. A medical test called coronary angiography with computed tomography (CTA) acquires 3D images of the heart and arteries and can provide doctors with an estimate of how narrow a patient’s arteries are. Until now, however, there was no simple, automated, and quick way to measure visible plaque in CTA images.
“Coronary plaque is often not measured because there is no fully automated way to do this“, he has declared Damini DeyPh.D., director of the quantitative image analysis laboratory at the Cedars-Sinai Biomedical Imaging Research Institute and senior author of the study: “When measured, an expert takes 25 to 30 minutes, but we can now use this program to quantify plaque from CTA images in five to six seconds. “
Dey and colleagues analyzed CTA images of 1,196 people undergoing coronary CTA at 11 sites in Australia, Germany, Japan, Scotland and the United States. Scientists trained the algorithm to measure plaque by practicing it through coronary CTA images donated by 921 people, which had already been analyzed by doctors specializing in the subject.
The algorithm is activated by first delineating the coronary arteries in 3-D images, then identifying the deposits of blood and plaque within the coronary arteries themselves. The investigators revealed that the instrument’s measurements matched the amounts of plaque seen in coronary AHUs and also compared the results with images acquired from two invasive tests considered highly accurate in assessing coronary artery plaque and narrowing: intravascular ultrasound and coronary angiography with catheter.
In a second step, the scientists found that measurements made by the AI (Artificial Intelligence) algorithm from CTA images accurately predicted the risk of heart attack within five years for 1,611 people who participated in a multicenter study called the SCOT-HEART study.
“More research is needed, but the possibility exists that we may be able to predict whether and how soon a person is likely to have a heart attack based on the amount and composition of plaque detected with this standard test “said Dey, who is also a professor of biomedical sciences at Cedars-Sinai.
Other research has used an algorithm to predict the onset of heart failure: “We have shown that deep learning algorithms can recognize blood pumping problems on both sides of the heart from the ECG waveform data “the assistant professor of genetics and genomics sciences said in a press release Benjamin S. Glicksberg.
“Usually, diagnosing this type of heart disease requires expensive and time-consuming procedures. We hope this algorithm will allow for faster diagnosis of heart failure“, The experts explained.
For years, doctors have relied on echocardiograms to assess whether a patient might be suffering from heart failure.. While this method is useful, it is also labor-intensive and not widely available.
“While interesting, traditionally it has been difficult for doctors to use ECGs to diagnose heart failure. This is partly due to the fact that there are no established diagnostic criteria for these assessments and because some changes in ECG readings are simply too subtle to be detected by the human eye.“, he has declared Girish N. Nadkarni, associate professor of medicine at the Icahn School of Medicine at Mount Sinai.
This study represents a “exciting step forward in the search for hidden information within ECG data that can lead to better screening and treatment paradigms using a relatively simple and widely available test ”.
The algorithm was 94% accurate in predicting which patients would experience a health ejection fraction and 87% in predicting those with less than 40% ejection fraction.. As for the interception of patients with a weakened heart, the algorithm had an accuracy of 73%.
The results also revealed that the algorithm learned to detect right valve weakness from EKGs with 84% accuracy. According to the researchers, the additional analysis could improve the ability of AI to detect cardiac weakness in all patients, regardless of race and gender.
“Our results suggest that this algorithm could be a useful tool to help clinical professionals fight heart failure suffered by a variety of patients, ”Glicksberg said. “We are carefully designing prospective trials to test their effectiveness in a more real environment“.
According to Dey and colleagues, it is important to continue to study how their algorithm is able to quantify plaque deposits in patients undergoing coronary CTA.
#Algorithm #predicts #heart #attacks