AI could lead to better ways to predict cardiovascular disease onset and progression

Scientists at the University of Utah Health have shown for the first time that artificial intelligence could lead to better ways to predict the onset and progression of cardiovascular disease. Researchers, in collaboration with doctors at Intermountain Primary Children’s Hospital, developed unique computational tools to precisely measure the synergistic effects of existing diseases on the heart and blood vessels.

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

Although the study only focused on cardiovascular disease, researchers believe it could have broader implications. In fact, they suggest that these findings could eventually lead to a new era of personalized, preventive medicine. Physicians would be proactive in contacting patients to alert them to possible discomfort and what can be done to alleviate the problem.

We can turn to AI to refine risk for virtually any medical diagnosis. The risk of cancer, the risk of thyroid surgery, the risk of diabetes – every medical term you can think of.”

Martin Tristani-Firouzi, MD study corresponding author and pediatric cardiologist at U of U Health and Intermountain Primary Children’s Hospital and investigator at the Nora Eccles Harrison Cardiovascular Research and Training Institute

The study appears in the online journal PLOS Digital Health.

According to Mark Yandell, Ph.D., senior author of the study, professor of human genetics, current methods for calculating the combined impact of various risk factors — such as demographics and medical history — on cardiovascular disease are often imprecise and subjective. HA and Edna Benning Presidential Endowed Chairs at U of U Health and co-founders of Backdrop Health. As a result, these methods fail to identify certain interactions that could have profound effects on heart and blood vessel health.

To more accurately measure how these interactions, also known as comorbidities, affect health, Tristani-Firouzi, Yandell and colleagues at U of U Health and Intermountain Primary Children’s Hospital used machine learning software to study more than 1.6 million Sort Electronic Health Records ( EHRs) after deleting names and other identifying information.

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

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

In adults, the researchers found the following:

  • People who had previously been diagnosed with cardiomyopathy (disease of the heart muscle) were 86 times more likely to need a heart transplant than those who had not.
  • Those who had viral myocarditis were about 60 times more likely to need a heart transplant.
  • The use of milrinone, a vasodilator drug used to treat heart failure, increased the risk of transplantation 175-fold. This was the strongest individual predictor for heart transplantation.

In some cases, the combined risk was even greater. For example, patients with cardiomyopathy taking milrinone were 405 times more likely to need a heart transplant than patients whose hearts were healthier.

According to Tristani-Firouzi, comorbidities had a significantly different impact on the risk of transplantation in children. Overall, depending on the underlying diagnosis, the risk of pediatric heart transplantation was 17 to 102 times greater than in children without pre-existing heart disease.

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

Infants who underwent Fontan’s surgery, a procedure that corrects a congenital blood flow defect in the heart, were about 20 times more likely to develop SND heart rate disorder than those who didn’t need the surgery

The researchers also discovered important 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 a similar medical history.

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

“This novel technology shows that we can accurately assess the risk of medical complications and even identify drugs that are better for individual patients,” says Bonkowsky.

In the future, Tristani-Firouzi and Yandell hope their research will also help doctors untangle the growing web of confusing medical information that surrounds them every day.

“No matter how conscious you are, there’s no way you can retain all the knowledge you need as a medical professional today to best treat patients,” says Yandell. “The computing machines we are developing will help doctors make the best possible decisions about patient care, using all the relevant information available in our electronic age. These machines are critical to the future of medicine.”


Magazine reference:

Wesołowski und S., et al. (2022) An explainable artificial intelligence approach to predicting cardiovascular outcomes using electronic medical records. PLOS Digital Health.

Leave a Reply

Your email address will not be published.