An AI-trained robot for human surgery could be trialled within a decade after its success on pig organs.
The robot, trained on videos of surgeries, carried out a lengthy phase of gallbladder removal without human assistance.
It operated for the first time on a lifelike patient, and during the operation, responded to and learned from voice commands from the team, similar to a novice surgeon working with a mentor.
Performing unflappably across trials, it demonstrated the expertise of a skilled human surgeon, even during unexpected scenarios typical in real-life medical emergencies.
The work, led by researchers from Johns Hopkins University in the US, represents a transformative advancement in surgical robotics.
Medical roboticist Axel Krieger said: ‘This advancement moves us from robots that can execute specific surgical tasks to robots that truly understand surgical procedures. 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.’
In 2022, Krieger’s Smart Tissue Autonomous Robot (STAR) carried out the first autonomous robotic surgery on a live animal – a laparoscopic operation on a pig.
But that robot required specially marked tissue, operated in a highly controlled environment and followed a rigid, predetermined surgical plan.
Krieger said it was like teaching a robot to drive along a carefully mapped route.
But this new system, he says, ‘is like teaching a robot to navigate any road, in any condition, responding intelligently to whatever it encounters’.
Surgical Robot Transformer-Hierarchy, SRT-H, can adapt to individual anatomical features in real-time, making swift decisions and self-correcting when things don’t go as expected.
Built with the same machine learning architecture that powers ChatGPT, SRT-H learns from feedback.
The robot responds to spoken commands such as ‘grab the gallbladder head’ and corrections like ‘move the left arm a bit to the left’.
Lead author Ji Woong ‘Brian’ Kim, a former postdoctoral researcher at Johns Hopkins who’s now with Stanford University, said: ‘Our work shows that AI models can be made reliable enough for surgical autonomy – something that once felt far-off but is now demonstrably viable.’
Last year, Krieger’s team used the system to train a robot to perform three foundational surgical tasks: manipulating a needle, lifting body tissue and suturing. Those tasks took just a few seconds each.
The robot had to identify specific ducts and arteries, grasp them accurately, strategically place clips, and cut parts with scissors. It performed flawlessly across varied anatomical conditions, even during unexpected detours, such as when the researchers changed the robot’s starting position or introduced blood-like dyes that altered the appearance of the gallbladder and surrounding tissues.
Johns Hopkins surgeon Jeff Jopling, a co-author, said: ‘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.’
Next, the team would like to train and test the system on a broader range of surgeries, expanding its capabilities to perform more comprehensive autonomous surgeries.
The findings are published in Science Robotics.
