Artificial intelligence identifies people at risk of heart complications

January 20, 2022 11:50 am

University of Utah Health scientists Mark Yandell, PhD, (left) and Martin Tristani-Firouzi, MD, have used artificial intelligence to better predict the onset and outcome of heart disease. Photo credit: Charlie Ehlertani-F

The system leverages electronic health records (EHRs) to assess combined effects of various risk factors

For the first time, scientists at the University of Utah Health have shown that artificial intelligence could lead to better ways to predict the onset and course of cardiovascular disease. The researchers, working in collaboration with physicians at Intermountain Primary Children’s Hospital, developed unique computational tools to accurately measure the synergistic effects of existing medical conditions on the heart and blood vessels.

The researchers say this comprehensive approach could help doctors predict, prevent or treat serious heart problems, perhaps even before a patient is aware of the underlying disease.

We can turn to AI to help refine risk for virtually any medical diagnosis

Martin Tristani-Firouzi, MD

Although the study only focused on cardiovascular disease, the researchers believe it could have much wider implications. In fact, they suggest that these discoveries could eventually lead to a new era of personalized preventive medicine. Doctors would proactively contact patients to alert them to potential ailments and what can be done to alleviate the problem.

“We can look to AI to help refine risk for virtually any medical diagnosis,” says Martin Tristani-Firouzi, MD, study corresponding author and pediatric cardiologist at U of U Health and U Health. ‘Intermountain Primary Children’s Hospital, and scientist at the Nora Eccles Institute for Cardiovascular Research and Training Harrison. “The risk of cancer, the risk of thyroid surgery, the risk of diabetes – any medical term you can imagine.”

The study appears in the online journal PLOS digital health.

Current methods for calculating the combined effects of various risk factors — such as demographics and medical history — on cardiovascular disease are often imprecise and subjective, according to Mark Yandell, Ph.D., lead study author, professor of Human Genetics, HA and Edna Benning Presidential Chair at U of U Health and co-founder of Backdrop Health. As a result, these methods fail to identify certain interactions that could have profound effects on heart and blood vessel health.

have cardiovascular
This figure depicts the University of Utah Electronic Health Records as a background network. The circles represent clinical variables such as health conditions, medical procedures, medications, and laboratory tests. Using artificial intelligence, U of U Health scientists scoured this database to find interactions between these variables to create a computational tool that could help doctors better predict risk factors for cardiac disease.

To more precisely measure the influence of these interactions, also called comorbidities, on health, Tristani-Firouzi, Yandell and their colleagues at U of U Health and Intermountain Primary Children’s Hospital used machine learning software to sort over 1.6 million electronic health records (EHRs) after names and other identifying information was removed.

These electronic records, which document everything that happens to a patient, including lab tests, diagnoses, medication use and medical procedures, have helped researchers identify the comorbidities most likely to worsen a condition. particular medical condition such as cardiovascular disease.

In their current study, the researchers used a form of artificial intelligence called probabilistic graphical networks (PGMs) to calculate how any combination of these comorbidities might influence the risks associated with heart transplants, congenital heart disease, or sinoatrial node dysfunction. (SND, a disturbance or failure of the natural pacemaker).

In adults, researchers found that:

  • People who had previously been diagnosed with cardiomyopathy (disease of the heart muscle) were at 86 times higher risk of needing a heart transplant than those who did not.
  • Those who had viral myocarditis had about a 60 times higher risk of needing a heart transplant.
  • Use of milrinone, a vasodilator drug used to treat heart failure, increased risk of transplant 175 times This was the best individual predictor of heart transplantation.

In some cases, the combined risk was even greater. For example, in patients who had cardiomyopathy and were taking milrinone, the risk of needing a heart transplant was 405 times higher than it was for those with healthier hearts.

Comorbidities had a significantly different influence on the risk of transplantation in children, according to Tristani-Firouzi. Overall, the risk of pediatric heart transplantation ranged from 17 to 102 times higher than that of children who had no preexisting heart problems, depending on the underlying diagnosis.

The researchers also looked at the influences that a mother’s health during pregnancy had on her children. Women who had high blood pressure during pregnancy were about twice as likely to give birth to babies who had congenital heart and circulatory problems. Children with Down syndrome were about three times more likely to have a heart defect.

Infants who had Fontan surgery, a procedure that corrects a congenital blood flow defect in the heart, were about 20 times more likely to develop SND heart rate dysfunction than those who did not need the surgery

The researchers also detected significant demographic differences. For example, a Hispanic patient with atrial fibrillation (rapid heartbeat) had twice the risk of SND compared to blacks and whites, who had similar medical histories.

Josh Bonkowsky, MD Ph.D., The director of the Center for Personalized Medicine for Primary Children, who is not one of the study’s authors, believes that this research could lead to the development of a practical clinical tool for patient care.

“This new technology demonstrates that we can accurately estimate the risk of medical complications and even determine which medications are best for each patient. said Bonkovsky.

Going forward, Tristani-Firouzi and Yandell hope their research will also help doctors untangle the growing web of disorienting medical information that envelops them every day.

“No matter how aware you are, there’s no way to keep all the knowledge you need in your head as a healthcare professional these days to treat patients in the best possible way,” says Yandell. “The computing machines we are developing will help physicians make the best possible patient care decisions, using all the relevant information available in our electronic age. These machines are vital for the future of medicine.


This research was published online on January 18, 2022 under the title “An Explainable Artificial Intelligence Approach to Predicting Cardiovascular Outcomes Using Electronic Health Records”.

In addition to Drs. S. Wesolowski, G. Lemmon, EJ Hernandez, A. Henrie, TA Miller, D. Wyhrauch, MD Puchalski, BE Bray, RU Shah, VG Deshmukh, R. Delaney, HJ Yost and K. Eilbeck.

The study was supported by the AHA’s Strategically Focused Children’s Research Network, the Nora Eccles Treadwell Foundation and the National Heart, Lung and Blood Institute.

Competing interests: Yandell, Deshmukh and Lemmon own shares in Backdrop Health; there is no financial link concerning this research.

Research news Cardiology Pediatric surgery Human genetics

Comments are closed.