MUHC inaugurates centre for surgical AI and robotics

MONTREAL – The McGill University Health Centre has launched SuPER – the Surgical Performance Enhancement and Robotics Centre at the Montreal General Hospital.

Its mission is to lead the way in research and development of AI-enabled technologies for surgical training, decision-support systems, robotics, and to become a centre of excellence not just in Canada but worldwide.

The launch is part of MUHC’s $35 million Future of Surgery strategy, an effort to improve procedures and outcomes for patients. The hospital has become a world leader in ERAS – Enhanced Recovery After Surgery – procedures that lead to better results by using best practices before, during and after surgical operations.

Similarly, the work on AI and surgical robotics is a key part of improving the patient experience. AI can help personalize surgical care before, during, and after surgery for each patient, while robot-assisted procedures can reduce the size of incisions and blood loss, leading to fewer infections, less pain and faster recovery times.

With cameras and surgical instruments attached to the ends of robotic arms, surgeons can work more accurately in tiny areas of the body, cutting, ablating, and sewing more efficiently than ever before.

As MUHC’s surgeon in chief, Dr. Liane Feldman, noted: “We are investing in next-generation technologies and analytics enabling smarter operative planning, data-driven training and augmented decision-making, helping to make surgery less invasive, safer and more personalized.”

There’s a great deal of excitement about the new SuPER centre, as it’s bringing together leaders in surgery, medical robotics and AI – all in a hospital setting.

For this reason, “SuPER is unique in Canada,” said Dr. Amir Hooshiar, founding director of the SuPER Centre, assistant professor of Surgery, and scientist with the Research Institute of MUHC.

He noted that Montreal is already a centre of excellence in artificial intelligence – with organizations like MILA and Google, as well as universities producing world-class innovations in AI – and it’s also a leader in minimally-invasive and robot-assisted surgeries. It’s combining these skills at the SuPER Centre.

“It’s based right at a major hospital,” Dr. Hooshiar added. “The centre is within walking distance of the operating rooms.”

In this way, SuPER can work closely with surgeons at MUHC, learning how they conduct procedures in abdominal surgery, orthopedics, oncology, cardiology, head and neck surgery, interventional radiology and other areas, and incorporating knowledge into the systems under development.

“We have world leaders in these areas, and they’re all working with SuPER,” said Dr. Hooshiar.

He noted that SuPER is addressing real-world issues, helping to improve procedures and outcomes by working directly with surgeons. “We’re addressing their needs and working with them to solve clinical problems.”

SuPER is part of the Research Institute of MUHC, and it’s located at the Montreal General Hospital. Already, it employs 20 staff and trainees at postdoctoral, doctoral, and graduate levels and plans to add another 10 trainees in the next six months.

To start up and sustain itself, SuPER is receiving critical support from the Montreal General Hospital Foundation.

It’s also working with industry partners, including Canadian start-ups and multinational companies. “We’re co-developing with industrial partners,” said Dr. Hooshiar. “We’re exchanging knowledge.”

The plan is to licence innovations to companies, so the technologies can reach patients in Canada and around the world.

An immediate focus is the development of the next generation of surgical robots, called “soft surgical robots”. Dr. Hooshiar explained these are different than traditional robots, which typically are capital-intensive and sport large and rigid arms.

In contrast, “Soft robots are modelled on the trunks of elephants and the arms of the octopus,” he said, explaining these structures are extremely flexible. When miniaturized, they’re able to gain access to many regions of the body that have previously been difficult to reach – such as regions behind the nose, for example.

In this case, surgeons will be able to snake small, soft robots through the nose and mouth, using cameras and instruments on the ends of the devices to conduct procedures, such as removing tumours.

Soft robots also have the ability to enter organs like the heart through major blood vessels, enabling the surgeons using them to conduct procedures with greater adroitness than ever before.

Dr. Hooshiar said different robots are needed for these new types of procedures, while the large robots produced by major vendors like Intuitive, are still extremely useful. In fact, an Intuitive da Vinci robot is currently located in the SuPER centre, said Dr. Hooshiar, where trainees are using it.

Of course, traditional surgical robots can be very expensive; another goal of the centre is to develop a generation of smaller robots that are more affordable for a greater range of hospitals.

Yet another stream of research at the centre is to develop inexpensive, disposable robotic devices. “You can mount them, then dispose of them,” he said.

He noted that different solutions are being arrived at by brainstorming with a wide variety of experts at the centre. “We have people with many different backgrounds, and they bring different ways of looking at things,” said Dr. Hooshiar.

The use of AI is a major thrust of the SuPER centre. One way that AI is being used is to model the way that surgeons conduct their procedures so these actions can be adopted by robots.

In this way, new iterations of surgical robots can incorporate the motions, nuances and behaviours of top surgeons, especially those with the best outcomes.

Dr. Hooshiar said that teams at SuPER will work on replicating these behaviours in AI models.

One way they’re doing this is through analyzing video captures of surgical operations, such as hernia repairs. “We have access to thousands of hours of surgical videos,” he said.

AI systems can be trained to model the actions of the surgeons – in hernia repair operations and dozens of other surgical procedures.

The knowledge can then be used in education and training systems, so that when medical students and residents conduct virtual operations the AI systems can observe them. “The AI can then tell them, you missed a step or you could have improved in a certain area.”

As well, the actions and behaviours can also be incorporated into surgical robots, so that some of the cognitive load in the OR is taken off the surgeon.

AI can act as an assistant to surgeons in the OR, automatically conducting actions or reminding surgeons of best practices. Moreover, computerized systems never get tired and can help ensure that all the steps in various surgeries have been taken.

Acting as a co-pilot in this way can result in less burnout of clinicians and improved outcomes for patients.

Dr. Hooshiar stressed that we’re a long way from surgical robots actually conducting procedures on their own.

Surgical operations are complex, involving many spur-of-the-moment decisions and collaborations with others in a team.

Human judgement and decision-making abilities are required that are beyond the capabilities of current AI systems – or even those that could be produced in the foreseeable future.

“It’s naïve to think that AI could replace surgeons,” he said. “The goal is to augment their skills in the OR and to improve patient outcomes.”