With AI, researchers can pixelate an image and use those characteristics to develop an algorithm, which can refine the data coming in from the image. That information can be used to help guide surgeons through the body during procedures.
One example he gave is a biotech company called On Target Laboratories Inc., which uses fluorescent markers discovered at Purdue University to target and illuminate cancer during surgery. Papay said the tech works by injecting a small particle that lights up with a specific light frequency into a patient with ovarian or colon cancer.
“It will light that cancer up like a lightbulb,” he said. “Then you can use artificial intelligence with the light frequencies that you’re reflecting on those particular parts of the body to know there’s definitely a tumor here and it goes and extends a little bit beyond that.”
Not a Modern Healthcare subscriber? Sign up today.
Another way AI can revolutionize surgical procedures is through robotics and machine vision, Papay said.
“Our spectral of light that we see by our human eyes is very limited,” he said. “It’s not limited by machine vision, by a robot … That can be used in a similar manner for artificial intelligence. To mine the data coming in from the robot’s machine vision to know what to cut and what not to cut (during surgery).”
There is also the possibility of utilizing AI to observe a surgeon during an operation.
Surgical Safety Technologies, a Toronto-based company, has developed a technology called the OR Black Box, which takes the concept of an aircraft black box to the operating room. The technology can be used to analyze what happens in an operating room environment. Using the OR Black Box, researchers can follow the outcomes of particular patients and look for correlations between their outcomes and what happened in the operating room, Papay said.
Ponsky also sees AI becoming a tool for measuring a surgeon’s capabilities.
Researchers at Case Western Reserve University were part of a team that developed a deep neural network to predict whether a person had undergone a rhinoplasty. The program was trained using a dataset of 22,686 before-and-after photos collected from publicly available sites, according to the study published in 2020. It correctly predicted rhinoplasty status in 85% of test images.
Download Modern Healthcare’s app to stay informed when industry news breaks.
A platform like this could be used to evaluate the quality of a surgeon’s work, Ponsky said.
Dr. Steven Williams, president of the American Society of Plastic Surgeons, said AI has the potential to improve efficiencies in the medical field, but that it can also have harmful effects on patients’ perceptions of themselves.
Williams, who is with California-based Tri Valley Plastic Surgery, noted that generative AI can create unrealistic images of people’s physical appearances. As a result, they may develop expectations for the results of a certain procedure that simply aren’t feasible.
“It’s going to be a double-edged sword because people are going to try and make themselves look like what they analytically should, but that’s not the definition of beauty,” Ponsky said. “Beauty is our uniqueness. We can’t just be computerized replicas of what the standard is.”
Despite the “slippery slope” that AI could create in regard to people’s perceptions, Ponsky said she believes advancements in AI will ultimately improve the medical field. It will serve as a tool for physicians, she said, and will allow for the use of more predictive factors, such as the likelihood of post-surgery reactions based on a patient’s medical history.
“(AI algorithms) are very powerful data processors,” she said. “They’re going to see these patterns that we humans cannot or don’t have the capacity for.”
This story first appeared in Crain's Cleveland Business.