A new test has been developed by researchers to predict at least 60 percent accuracy in the possibility of a heart attack within four years. According to a study published in a peer-reviewed journal, the test used machine learning and artificial intelligence to analyse “biomarkers” of heart disease found in least 27 proteins in the blood.
The AI-based method utilises dozens of proteins found in blood as markers to detect heart attack. The test is about 60 percent more accurate than current 'score' system in the projection of likelihood of heart attack.
Termed 'Protein model' test uses at least 27 blood-borne proteins to uncover risk of cardiac event. The score system helps in establishing overall risk of heart disease among individuals. Some of the tools included in the system are cholesterol test, blood pressure, lifestyle factors, family history and health conditions such as diabetes.
Prediction of heart attack with greater accuracy
It may be possible using the new test to predict the possibility of a heart attack. According to reports, the test uses artificial intelligence (AI) to analyse a host of proteins in the blood to provide a much higher accuracy in predicting cardiovascular events the body may be facing.
Compared to the existing 'risk score' system, the protein-based analyser has 60 percent more accuracy in predicting a risk of heart attack within a period of 48 months.
What is the new test?
The protein model uses at least 27 proteins found in the blood. The number of significant as it is about a quarter of all proteins encoded by the genes.
Why only now?
Thanks to emerging new technologies and improving computing power, the method has become available in current circumstances. It allows experts to measure thousands of blood proteins to assess patients' individual risk.
Who invented it?
The test has been developed by researchers led by Dr Stephen Williams, based in Colorado. They published their work in Science Translational Medicine journal on April 6, 2022.
Dr Williams, the chief medical officer at US-based SomaLogic, is a specialist in translational medicine and undertakes the management of a clinical assay laboratory. His research team included 22 experts in different medical disciplines from various parts of the world to work together to develop the test.
SomaLogic’s test measures protein to categorise people between high and low risk groups, along with providing a percentage likelihood of a cardiovascular event within the next four years.
How many bloods samples were used to test its accuracy?
The protein method used 32,130 “archived” plasma (amber-colored liquid in the blood) samples from 22,849 participants in nine clinical studies to analyse the accuracy of the test.
According to the journal, Dr. Williams used machine-learning to analyse 5,000 proteins in blood plasma samples from 22,849 people. This method helped them in identifying a signature of 27 proteins that helped in the prediction of likelihood of heart attack, stroke, heart failure or death, over a period of 48 months.
What is the role of blood proteins in predicting heart attacks?
In general, heart disease or cardiovascular risk factors can activate several pro-oxidative genes in the blood vessel wall. This generates reactive oxygen species, especially those generated by the endothelium - a thin membrane that lines the inside of the heart and blood vessels.
What is the role of the endothelium in the production of these proteins?
In medical science, endothelial cells release substances that control vascular relaxation and contraction. They also release enzymes that control blood clotting, immune function and platelet adhesion. These genes results in a complex cascade of events towards the transition from normal endothelial function to dysfunction.
It further results in abnormal heart activity and the development of a pro-coagulant endothelial surface, inflammation and plaque formation. This combination further blocks the flow of oxygenated blood to the heart, leading to a heart attack.
What is the significance of this new test?
The risk of heart disease is increasing globally despite various treatments. According to a 2012 study published in the journal JAMA, the risk of developing heart disease is more than 40 percent in men and 30 percent women over a lifetime.
Can you really predict when a heart attack would strike?
According to Dr Sutter Zi-Jian Xu, a cardiologist in the Sutter Health network, there are subtle heart attack symptoms months before the actual event takes place. The existing risk score system undertakes a person’s age, sex, race, medical history, cholesterol and blood pressure to analyse the possibility of a cardiovascular event.
The new 27-protein model enhances the benefits of scoring method. In comparsion, the new test accurately assessed the risk in people who have previously had a heart attack or stroke or have additional illnesses or are taking medicines to reduce their risk.
Where does the ‘60% accuracy’ taken from?
In the study, researchers underlined that less than a quarter of people testing positive had a heart attack, which is equal to positive predictive value of 22% (at 95% confidence interval).
The test further identified a majority of patients who died or had a major cardiac adverse event in the past 30 days of the study (98.0% sensitivity).
The protein model test is different from genetic tests as the former provides a more accurate picture of the activities of a person's organs, tissues and cells at any given moment in time, William’s team explained.
Advantage of the protein test
One of the most significant predictor for heart attack risk is having either high LDL cholesterol (bad cholesterol) or low HDL cholesterol (good cholesterol) or both.
Genetic tests, on the other hand, can provide an idea of someone’s risk of certain diseases on the basis of their genetic profile.
Facts about heart disease
Heart disease is one of the most prominent causes of death for men, women, and people of most racial and ethnic groups in the US.
As per the data, one person dies every 36 seconds in the United States from cardiovascular disease. Furthermore, about 659,000 people die from heart disease in the US each year—1 in every 4 deaths.
How much is this test? When can I get it?
According to Dr Williams, the test is being used in four healthcare systems within the US, however, the price is not immediately available.
Speaking to the media, Prof Manuel Mayr, the British Heart Foundation professor of cardiovascular proteomics at King’s College London, stated that proteins are the building blocks of the body. While the study has uncovered new links between proteins in blood and death by all causes, there is need for further research to assess the potential clinical impact of using these 27 proteins as prediction tools for cardiovascular disease.
SOURCE: Gulf News