The Daily Pennsylvanian is a student-run nonprofit.

Please support us by disabling your ad blocker on our site.


Penn Medicine is studying how to use AI to recognize if patients’ bodies will accept a newly transplanted heart.

Credit: Kylie Cooper

Penn Medicine received a $3.2 million grant to study the use of artificial intelligence in improving heart transplant outcomes.

The National Institutes of Health provided the grant which Penn Med will use to fund a project to test whether AI analyses can help determine whether a patient will accept or reject a heart transplant, Penn Medicine News reported

One of the greatest risks of a heart transplantation is the possibility that the patient’s body may consider the new heart to be a foreign object, Penn Medicine News reported. In such a case, the immune response from the body can damage the heart, leading to life-threatening consequences for the patient. According to NIH, these rejections are responsible for 10 percent of all deaths within three years of transplant surgery. 

Determining if a patient's body will reject a heart can be difficult, and current metrics for organ rejection can be unreliable and inaccurate, Penn Medicine News reported.

Kenneth B. Margulies, professor of Cardiovascular Medicine at Penn Med, is leading the effort to train computers to recognize if patients’ bodies will accept a newly transplanted heart. Michael D. Feldman and Priti Lal from the Department of Pathology and Laboratory Medicine and Eliot Peyster from Cardiovascular Medicine will assist Margulies, according to Penn Medicine News.

Margulies and his team are working closely with Anant Madabhushi, a researcher at Case Western Reserve University, to make use of the data, which consist of digitized images of patients after a biopsy. With the help of AI, computers will analyze the images and determine the probability that the patient’s body will either reject or accept the heart. 

In addition to improving patient outcomes, one of the goals of the research is to compare the performance of the AI technology to that of human pathologists. Previous studies have found computers are more effective at diagnosis, but Madabhushi and other researchers believe the technology will be a tool for human specialists, not a replacement, Penn Medicine News reported. 

“This research is focused on a critical component of heart transplantation—improving patient outcomes," Margulies told Penn Medicine News. "Unfortunately, the number of patients with end-stage heart failure is increasing. But research like this is another step in the right direction for improving survival and quality of life for heart failure patients."