When Dr. Amin Madani does not remove ruptured appendixes or excise cancer cells from his patients, he thinks about how he can improve surgeons’ performance in the operating room.
This is because up to 25 percent of the millions of people who undergo inpatient surgery around the world each year experience negative complications either during or after surgery, according to the World Health Organization. These adverse effects can range from tenderness at the incision site to internal bleeding to death.
While not all of these adverse events are caused by surgeons’ actions, some are, and Madani – a general surgeon with the Sprott Department of Surgery at the University Health Network (UHN) in Toronto – wants to reduce this risk.
He researched the techniques and thought processes used by the most elite, highly qualified surgeons, when a group of computer and computer scientists suggested that he could use artificial intelligence (AI) to mimic their minds.
“I was actually a big skeptic for the longest time,” Madani said. “It’s a great statement to make.”
The resulting collaboration produced a prototype that uses computer vision – a field of AI that trains computers to interpret and understand images – to identify in real time areas of a body that are safe to dissect and those where it is dangerous to do it.
SE | How artificial intelligence identifies safe dissection areas during gallbladder surgery:
It is part of a barrage of activity in recent years among researchers, healthcare professionals and companies trying to harness the power of digital technology to provide better medical care.
Madani’s technology is still in its early stages and currently only applicable to gallbladder surgeries. But he says it has the potential to improve surgery around the world, especially in rural, remote areas and lower-income countries that lack surgical expertise.
Other experts agree, though they say there are still challenges to overcome before its potential can be realized.
How technology helps guide surgeons
When surgeons perform a gallbladder operation, they make a “keyhole incision” in the patient’s abdomen, insert a camera into the abdomen, and use special tools to cut away and remove the organ.
Madani’s technology projects colored areas on the video monitor that the surgeon uses to see inside the patient’s body. Green means that the area of the organ is safe to cut, red means that it is not.
Another iteration uses a heatmap-like projection that changes color based on the model’s confidence in where the safe area is.
SE | Dr. Amin Madani explains how the prototype could help guide surgeons:
The prototype was developed by feeding hundreds of hours of videos of gallbladder surgeries into a software program and integrating comments from expert surgeons who identified where they would dissect. After analyzing the data frame by frame, the algorithm begins to recognize patterns and develops the ability to make independent decisions.
The algorithm was able to consistently identify “go” and “no-go” zones as well as the liver, gallbladder and hepatocystic triangle with an accuracy of between 93 and 95 percent, according to a 2020 study of 290 videos from 153 surgeons who were published in the academic journal Annals of Surgery. Madani was the lead author.
“It’s like I have a panel of experts standing and watching over my shoulder, guiding me, navigating me and helping me not get into trouble during that surgery,” Madani said.
Dr. Daniel Hashimoto, an operations instructor at University Hospitals and Case Western Reserve University in Cleveland, Oh., Who collaborated with Madani on the study, said the technology’s real promise lies in its ability to help surgeons better understand what they perceive. when making surgical decisions.
“The hope is to say, well, can we bring another pair of eyes into the operating room – in this case machine eyes – to make sure the surgeon sees what they think they see?” said Hashimoto.
The next question is: will it actually improve surgeons’ performance in the operating room and reduce complications?
It is a difficult question to answer from a research perspective, according to Hashimoto, because clinical trials studying side effects require a large number of patients to participate. But Madani is determined to find out.
His team has already tested the prototype during live operations to ensure it works properly, and now they are seeking approval from the UHN Ethics Council to conduct further research.
AI needs more data, videos
Another challenge lies in extending the technology to other surgical procedures.
Gallbladder surgeries are one of the most common surgeries, so it was relatively easy for the researchers to obtain videos of successful surgeries. However, tracking useful videos can become more difficult with less common procedures.
“Most modern machine learning basically depends entirely on the data,” said Frank Rudzicz, a professor of computer science at the University of Toronto and an expert in artificial intelligence in health care.
“If it has very few examples of something, it just will not learn the properties of that thing, and it will … perform very poorly.”
SE | The challenges of integrating artificial intelligence into healthcare:
Rudzicz said another challenge is designing the technology so that it increases the surgeon’s performance without causing distraction.
“One thing we do not want is for the surgeon to watch this video and everything lights up like a Christmas tree,” Rudzicz said.
Madani said he is well aware of the need for data and is already in talks with other experts to create a global repository of surgical videos.
Next, pending research approval, Madani plans to test whether the technology actually improves the performance of other surgeons, thus reducing negative surgical complications.