Few developing technologies have garnered as much attention and ignited the imagination of surgeons in recent years as artificial intelligence (AI).
In today’s Panel Session at 2:30 pm, Next Generation Predictive Tools and Artificial Intelligence for Anticipating Postoperative Outcomes, surgeons and national experts in biomedical AI will discuss how the technology can be harnessed to improve patient outcomes.
This session is especially timely because of the staggering growth of AI in the medical space, according to moderator Christopher J. Tignanelli, MD, FACS, associate professor in the Division of Critical Care/Acute Care Surgery at the University of Minnesota in Minneapolis. He explained that there are now more than 500 US Food and Drug Administration-approved AI solutions on the medical market.
“As a surgeon, you are going to see more and more AI show up in your practice, whether it’s the vendors that you work with who will have AI-embedded solutions or your hospital wanting to add some AI-based software,” said Dr. Tignanelli, who also is the co-director for the University of Minnesota Center for Clinical Quality & Outcomes Discovery and Evaluation, as well as the associate dean for data science.
Because of AI’s increasing prevalence, it is important for clinicians, practicing surgeons, and surgeon researchers to understand what some of the technology’s risks and benefits are, what it can do and what it cannot do, and how bias affects AI models.
“And, importantly, surgeons should know what some examples are of how AI is being used currently in surgery—in the preoperative setting, intraoperative setting, and postoperative setting,” Dr. Tignanelli said.
Preoperative models, which look at ways AI can help outline postoperative risks for morbidity and mortality, will be discussed and include a description of the ACS National Surgical Quality Improvement Program® risk calculator—a tool that almost all surgeons have used at least once, Dr. Tignanelli said. Alternative platforms such as MySurgeryRisk and the Predictive OpTimal Trees in Emergency Surgery Risk (POTTER) calculator also will be part of the conversation.
The discussion on the intraoperative potential will focus on future application of AI, including how tools based on data from hundreds of previous cases could help “predict” upcoming portions of a surgery while it is taking place and provide additional oversight, as well as the potential development of autonomous robot surgery. Models that can optimize the care of a patient by continued analysis and prediction of complications and other outcomes after patients are in recovery will be examined, too.
Dr. Tignanelli said he believes that the constantly evolving technology and its broad potential could lead to a robust discussion.
“Since the advent of ChatGPT and other popular AI platforms, I’ve received many requests for more information on these tools,” he said, “so I’m anticipating a great turnout for this session.”
