Sept. 10 (UPI) -- Artificial intelligence can help screen adults for heart disease by detecting specific bacteria in the gut, according to a study published Thursday by the journal Hypertension.
The researchers analyzed stool samples collected from nearly 1,000 patients -- half of whom had known heart disease -- with their state-of-the-art machine learning modeling tool that runs on artificial intelligence, or AI.
The approach correctly identified clusters of gut bacteria present in those with heart disease, researchers said.
Based on the presence of this bacteria, the AI-based tool correctly identified people with heart disease more than 70% of the time.
Related
"Gut microbiota are associated with the development of hypertension, or high blood pressure, 'the silent killer' that often leads to heart disease," study co-author Bina Joe told UPI.
"Our study shows that there are real changes in the population of these bacteria in our gut that can pinpoint people at risk for heart disease, before hypertension develops," said Joe, professor and chair of the department of physiology and pharmacology at the University of Toledo.
Nearly half of all U.S. adults have some form of heart disease, and many of them go undiagnosed until they suffer a major event like a heart attack or stroke, according to the American Heart Association.
The new AI-based screening tool could help enhance screening for heart diseases by pinpointing the presence of specific bacteria in the gut, Joe said.
The tool uses historical data on the presence of certain bacteria in patients with high blood pressure or heart disease to identify those at risk for the conditions -- based on whether they too have the same bacteria -- she said.
The gut microbiome is made up of microorganisms, including bacteria, that live in the digestive tracts of humans and are involved in the digestion of food and processing of nutrients.
In recent years, bacteria in the gut has been linked with overall health beyond the digestive tract, Joe said. Recent studies have also suggested that the presence of certain bacteria can increase risk for heart disease.
Because the composition of the bacteria in the gut varies from person to person, it has been challenging to develop a test that accurately identifies the bacteria that increases heart disease risk.
The AI-based tool, however, still is in the early stages of development, and it may be several years before doctors can use it to identify at-risk patients, Joe said.
"The gut microbiome is highly variable among individuals, so we were surprised by the promising level of accuracy obtained from these preliminary results," she said.
"It is conceivable that clinicians could analyze the gut microbiome of patients' stool samples with our machine learning method to screen patients for heart and vascular diseases."
