Imma Perfetto
Cosmos science journalist
Painful gallstones and recurrent infections mean that, sometimes, your pesky gallbladder has to go. The minor laparoscopic (keyhole) surgery is performed by a surgeon today, but one day these and other soft tissue surgeries could be the domain of robots.
A robot trained on videos of surgeries has performed a lengthy phase of a gallbladder removal, or cholyecystectomy, almost autonomously. This was done on a lifelike model – the gallbladder of a pig cadaver.
During the operation, the robot responded to and learned from voice commands given by a surgical team.
The findings are published in a study in Science Robotics.
“This advancement moves us from robots that can execute specific surgical tasks to robots that truly understand surgical procedures,” says Axel Krieger, a medical roboticist and associate professor at Johns Hopkins University in the US. “This is a critical distinction that brings us significantly closer to clinically viable autonomous surgical systems that can work in the messy, unpredictable reality of actual patient care.”
Cosmos reported on the Smart Tissue Autonomous Robot (STAR) when it was unveiled in 2016. It performed a supervised intestinal anastomosis, a surgery where 2 segments of bowel are stitched together. This was done in a live pig, but the feat required a large incision to access the intestine and significant guidance from humans.
Pigs are useful stand-in for the human body in preclinical research because their organs are of a similar shape and size.
In 2022, an updated STAR iteration performed the surgery in 4 live pigs using a laparoscopic approach. However, it required specially marked tissue, operated in a highly controlled environment, and followed a rigid, predetermined surgical plan.
The latest iteration of the system uses a “Surgical Robot Transformer-Hierarchy” (SRT-H) framework to adapt to individual anatomical features in real-time, make decisions on the fly, and self-correct when things don’t go as expected.
Gallbladder removal surgery performed by a robot. Credit: Johns Hopkins University
SRT-H is interactive too. It can respond to spoken commands – “grab the gallbladder head” – and corrections – “move the left arm a bit to the left” – and learns from this feedback.
It was trained through “imitation learning” on videos of Johns Hopkins surgeons doing gallbladder removals on pig cadavers. The visual training was reinforced with captions describing the tasks.
The robot carried out the clipping and cutting portion of a cholecystectomy. This string of 17 tasks involved grabbing the gallbladder, identifying ducts and arteries, strategically placing clips and severing parts with scissors.
It performed the surgery on 8 pig gallbladders with 100% accuracy but took longer to do so than a human surgeon.
“Just as surgical residents often master different parts of an operation at different rates, this work illustrates the promise of developing autonomous robotic systems in a similarly modular and progressive manner,” says co-author and Johns Hopkins surgeon, Dr Jeff Jopling.
It even adapted to unexpected conditions, such as an altered starting position and blood-like dyes which changed the appearance of the gallbladder and surrounding tissues.
“To me it really shows that it’s possible to perform complex surgical procedures autonomously,” Krieger said. “This is a proof of concept that it’s possible and this imitation learning framework can automate such complex procedure with such a high degree of robustness.”
The team now plans to train the system on more types of surgeries and expand its capabilities to perform these completely autonomously.