Computers “learn” to take charge of high-stakes decisions during heart surgery


VA experts in artificial intelligence teach computers to think like “perfusionists” – the operating room staff who use heart-lung machines during surgery.

Doctors often stop a patient’s heart during major heart surgery. In this situation, a perfusionist uses a heart-lung bypass machine. The machine supplies the necessary blood and oxygen to the patient whose heart is paused.

In an article titled “Using Machine Learning to Predict Perfusionists’ Critical Decision-Making during Cardiac Surgery,” researchers at the VA Boston Healthcare System, Harvard Medical School, Brigham and Women’s Hospital, and Georgia Tech University detail their progress towards a computer system system to help perfusionists in the operating room make critical decisions when using the bypass machine.

They presented the document in September 2021 at a meeting of the Medical Imaging Computing and Computer Assisted Intervention Society.

Support perfusionists when the pressure is high

Within VA and other healthcare facilities across the country, the past few decades have seen sophisticated technology enter the operating room. And a growing number of medical specialists are harmonizing their roles in the care of patients. While these systems have improved patient health, they can also increase pressure on operating room staff.

In this pressure cooker environment, the perfusionist must make quick and precise decisions in the management of the extracorporeal circulation (cardiopulmonary) machine, explains Dr. Marco A. Zenati, principal investigator of the study. Zenati is the Division Chief of Cardiac Surgery at the VA Boston Healthcare System and Professor of Surgery at Harvard Medical School.

Even expert perfusionists can be overwhelmed, given the demands placed on them, explains the physician-researcher, who heads the laboratory for medical robotics and computer-assisted surgery at the Boston VA and Harvard. A computerized system in partnership with perfusionists can improve patient safety, he says. This can potentially prevent unwanted events such as kidney damage or a blood clot due to heart-lung bypass surgery.

Prepare computers to pick up models

Dr. Lauren R. Kennedy-Metz, article author and surgical instructor at Harvard Medical School, also emphasizes the importance of the project. “The patient is at the mercy of the cardiopulmonary bypass machine during cardiac surgery, and the perfusionist is under great pressure to make decisions regarding the use of this very complex machine,” says Kennedy-Metz, member of the VA-Harvard laboratory at Zenati. “The goal of this project is to apply a form of artificial intelligence called machine learning to make recommendations that perfusionists can draw on with their own expertise.”

Veterans are among those who will benefit the most. They can be particularly susceptible, Kennedy-Metz explains, to adverse events during heart surgery, as they are often older and suffer from conditions such as diabetes. Kennedy-Metz says she is proud of VA’s contribution to this study, including its role as the site for all data collection. She cites her VA site in Boston, where the study was based, as a prime example of leadership not only in this study, but in “a constant stream of very impactful research.”

One of the responsibilities of perfusionists is to monitor the delivery of oxygen (DO2) while the patient is connected to the shunt device. When2 falls below the threshold of 280 ml / min / m2, the patient is at risk of complications. The study focused on the responses of perfusionists to these dangerous drops in DO2. The goal is to allow computers to make decisions like master perfusionists, but without the constraint that can undermine human judgment.

More information

Click on here to read the full story.

Click on here to learn more about VA research.