The 'robot doctor' operates alone, the first surgery without human assistance.

As unfazed as an expert doctor. A robot has performed the first surgery without human assistance . Trained with videos of operations, it autonomously completed the complex gallbladder removal procedure, operating for the first time on a real patient . During the procedure, it responded to and learned from the team's voice commands, like a novice green-coat surgeon working with a mentor. Even in unexpected scenarios typical of medical emergencies, the robot displayed the same expertise as a skilled human surgeon. The feat was reported in 'Science Robotics'.

The work led by researchers at Johns Hopkins University—and funded by U.S. federal funds—represents a major advance in surgical robotics, allowing robots to operate with machine-like precision while also offering human-like adaptability and understanding . It's a leap that "takes us from robots capable of performing specific surgical tasks to robots that truly understand surgical procedures," says associate professor and medical roboticist Axel Krieger. A "fundamental distinction," he clarifies, "that brings us significantly closer to clinically viable autonomous surgical systems, capable of operating in the chaotic and unpredictable reality of actual patient care."

In 2022, Krieger's STAR (Smart Tissue Autonomous Robot) robot performed the first autonomous robotic surgery on a living animal: a laparoscopic procedure on a pig. But that robot required specially labeled tissue, operated in a highly controlled environment, and followed a rigid, predetermined surgical plan. For Krieger, it was like teaching a robot to drive along a carefully plotted path. With the new system, however, "it's like teaching a robot to navigate any road, in any conditions, responding intelligently to everything it encounters," he says. The new surgical robot is called Transformer-Hierarchy, or Srt-H, and it truly performs surgeries by adapting to individual anatomical characteristics in real time, making decisions on the fly and self-correcting when things don't go as planned.

Built with the same machine learning architecture that powers ChatGpt , 'Dr. Srt-H' is also interactive, able to respond to voice commands ("grab the head of the gallbladder") and corrections ("move the left arm slightly to the left"), and learn from the feedback it receives. This overcomes "some of the fundamental obstacles to implementing autonomous surgical robots in the real world," emphasizes lead author Ji Woong "Brian" Kim, a former postdoctoral researcher at Johns Hopkins, now at Stanford University. "Our work demonstrates that AI models can be made reliable enough for surgical autonomy, a goal that once seemed distant but is now achievable," he assures.

Last year, Krieger's team used the system to train a robot to perform three basic surgical tasks: manipulating a needle, lifting tissue, and suturing. Each of these tasks took only a few seconds. The gallbladder removal procedure is much more complex, consisting of a sequence of 17 actions lasting several minutes . The robot had to identify and precisely grasp specific ducts and arteries, strategically place clips, and sever certain sections with scissors. Srt-H learned to perform the operation by watching videos of Johns Hopkins surgeons. The team reinforced the visual training with captions describing the tasks. After watching the videos, the robot performed the procedure with 100% accuracy. Although it took longer than a human surgeon to complete the procedure, the results were comparable to those of an experienced surgeon.

"Just as surgical residents often master different parts of an operation at different paces, this work illustrates the promise of developing autonomous robotic systems in a similarly modular and progressive manner," says Johns Hopkins surgeon Jeff Jopling, a co-author of the study. At one point, to test it, the researchers also adjusted the robot's starting position and added blood-like dyes that changed the appearance of the gallbladder and surrounding tissue. "To me, this really demonstrates that it's possible to perform complex surgical procedures autonomously," concludes Krieger. The team now plans to train and test the system on more types of surgeries and expand its capabilities to perform fully autonomous surgery.