CIOTechie_ArtificialIntelligence_Heart-attack_Researchers

Researchers Explore Artificial Intelligence To Predict Heart Attacks

Artificial Intelligence News

Researchers_Explore-Artificial-Intelligence-To-Predict-Heart-Attacks Researchers Explore Artificial Intelligence To Predict Heart AttacksAccording to the Center for Disease Control and Prevention, more than 610,000 people die due to heart disease every year, which is the leading cause of death for both men and women in the U.S. Thus for taking care of that, scientists at IBM and AstraZeneca, pharmaceutical giant have explored a Machine Learning solution that can identify early ACS (Acute Coronary Syndrome) warning signs. In ACS blood flow to the heart is abruptly obstructed and can lead to a heart attack if not treated on time.

To combat this condition, the scientists sourced a set of data including the age, gender, habits, personal disease history, procedures, laboratory test results, ACS type, and around 40 other traits of 26,986 adult hospitalized patients across 38 urban and rural hospitals in China, which they fed to a neural network, layers of mathematical functions loosely modeled after biological neurons. As per said neural network was formed to foresee the four factors at the same time, includes whether they’d experienced a Major Adverse Cardiac Event (MACE), prior to ACS; whether they’d obtained antiplatelet medicine to thwart blood clots from making in the coronary arteries; whether they’d been given beta-blockers that reduce blood pressure; and whether they were prescribed statins, a flocks of drugs that support lower cholesterol levels and in turn put off heart attacks and stroke.

After that, the researcher implemented k-means clustering, a statistical method in which data points are allocated to collections by similarities to put in order the patients into seven groups based on the classification data received from the neural network. From these, they collected constructive insights, in the first cluster, in which they grouped people with high MACE rates but low rates of disease, diabetes was one of the most ACS predictors, while in another cluster this one containing patients with severe conditions, age, and systolic blood pressure played an outsized role in ACS progression. Though the clustering has implication for diagnosis, it is not clear whether it can notify clinical practice. But also they said their effort showed that AI-powered cluster analysis is a promising approach for ACS patients.