Navigation auf uzh.ch
Navigation auf uzh.ch
Surgeons use all their senses and draw on their experience during complex operations. When they don’t have a clear view of a patient’s anatomy, they use tactile senses to assess the situation or listen for acoustic clues to determine, for example, when they should stop drilling. Surgical robots, however, lack these human abilities and nuanced senses, which significantly limits their range of use. Similar to an autopilot, these robots follow a predefined path based only on medical imaging. If this path does not accurately reflect the actual circumstances, the systems lack the necessary non-visual sensory perception needed to adapt, requiring the surgeon to take over again. Such challenges often arise during complex operations that demand high precision. These include, for example, spinal fusions where multiple screws must be inserted in close proximity to vital structures with pinpoint accuracy.
As part of the Horizon 2020 research program FAROS (Functionally Accurate RObotic Surgery), an international research consortium including Balgrist University Hospital and UZH has now developed a prototype of a surgical robot capable of autonomously performing complex and highly precise surgical tasks using a range of sensory perceptions mimicking human senses. It learns how to scan, touch, hear, feel and interpret the tissue, enabling it to respond accurately.
So, does the newly developed robot surgeon already have what it takes to master complex clinical and surgical situations? FAROS researcher Mazda Farshad, medical director at Balgrist University Hospital and professor of medicine at UZH, is in an excellent position to answer this question.
“We’ve developed a research prototype which can’t yet be directly used without validation in everyday hospital operations,” says Mazda Farshad “We need to carry out a few validation steps and safety assessments in the context of studies. The system wouldn’t yet function reliably enough on patients who have an exceptionally complex anatomy.”
What are the benefits of robotic assistance systems in spine surgery compared to human surgeons? “So far, there’s no clinically relevant added value with the existing systems,” says Farshad. “We need smart robotic systems that create real added value, especially when the outcome is highly dependent on surgical talent and skill. Only then would it be possible to increase the average surgical quality.”
Does this mean that spine surgeons’ role will shift to monitoring the work of surgical robots in the future? “Perhaps for certain surgical tasks, but not for a surgery as a whole,” says Farshad. “Ordinary surgical interventions are made up of around 300 sub-steps with many variables. A robot could probably take over a few of these steps, but not all of them.”
The FAROS project’s principal investigator was Philipp Fürnstahl, professor of orthopedic computer science at UZH and head of the Balgrist ROCS – Research in Orthopedic Computer Science group at Balgrist University Hospital.
The Swiss team worked hand in hand with Sorbonne University (France), King’s College London (UK), KU Leuven (Belgium) and an industry partner in France. The project was supported by the EU’s Horizon 2020 research and innovation program. “The project was an excellent experience,” says Philipp Fürnstahl. “The Horizon 2020 program is highly competitive, which is why the scientific level is very high. Our partners’ skills complemented ours very well, and the researchers were able to learn and benefit a lot from each other.”
What was the biggest surprise for him? “Even though researchers came from adjacent fields such as robotics, sensory technology or computer-assisted surgery, it took longer than expected for us to find a shared specialist language in all areas,” says Fürnstahl.
According to the UZH professor, the biggest challenge was the project’s final phase, when it came to combining the new robotic parts developed by the different research groups in Belgium, France and England. “This meant we had to fly in the whole research team and their equipment for so-called integration weeks so that we could assemble and validate a functioning robot in a short time,” recalls Fürnstahl. After three intense years and eight joint integration weeks, the final robotic prototype was tested at the Balgrist OR-X research and training center.
