A new robotic system has performed the first lifelike surgery without human assistance, marking a significant leap in the field of automated medical procedures.
In a remarkable leap for medical technology, a surgical robot has successfully conducted a gallbladder removal procedure on a lifelike patient model. This autonomous operation marks the first time a robot has performed such a complex task without direct human intervention, responding to voice commands and adapting in real time.
This federally funded work, led by Johns Hopkins University researchers and published in the latest edition of Science Robotics, marks a significant breakthrough in surgical robotics, enabling robots to combine mechanical precision with human-like adaptability and understanding.
Surgical Precision Meets Human-Like Adaptability
“This advancement moves us from robots that can execute specific surgical tasks to robots that truly understand surgical procedures,” co-corresponding author Axel Krieger, a medical roboticist at Johns Hopkins, said in a news release. “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 previous creation, the Smart Tissue Autonomous Robot (STAR), achieved a milestone by performing autonomous laparoscopic surgery on a live animal.
However, the new Surgical Robot Transformer-Hierarchy (SRT-H) is a milestone ahead, equipped with a sophisticated machine learning system that allows it to adapt to various surgical scenarios, akin to a human surgeon navigating the complexities of patient care.
A Leap Toward Fully Autonomous Surgery
SRT-H was built using the same machine learning architecture behind ChatGPT, enabling it to interact with surgical teams through spoken commands and feedback.
During the procedure, the robot displayed an impressive capacity to respond to corrections and make decisions autonomously.
“This work represents a major leap from prior efforts because it tackles some of the fundamental barriers to deploying autonomous surgical robots in the real world,” added lead author Ji Woong “Brian” Kim, a former postdoctoral researcher at Johns Hopkins who’s now with Stanford University. “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.”
From Basic Tasks to Complex Procedures
Building on last year’s success, where the team trained a robot to execute foundational surgical tasks, SRT-H has advanced to conducting a full-length gallbladder removal procedure, consisting of 17 specific tasks. This involved identifying and manipulating ducts and arteries, placing clips and using scissors with precision.
The robot was trained by watching videos of Johns Hopkins surgeons performing the surgery on pig cadavers, paired with descriptive captions to enhance learning.
Impressively, the robot achieved a 100% accuracy rate in its performance, although it took more time than a seasoned human surgeon.
Future Prospects and Continuing Development
The robot excelled in handling variable anatomical conditions and unexpected challenges introduced by the researchers, demonstrating its robustness and adaptability.
Johns Hopkins surgeon Jeff Jopling, a co-author of the study, emphasized the modular and progressive learning framework, likening it to how human surgical residents master complex skills over time.
“To me, it really shows that it’s possible to perform complex surgical procedures autonomously,” Krieger added. “This is a proof of concept that it’s possible and this imitation learning framework can automate such complex procedures with such a high degree of robustness.”
Looking ahead, the team aims to expand the robot’s capabilities to perform a broader spectrum of surgeries and move closer to achieving a fully autonomous surgical system.
The study was co-authored by a diverse team of researchers from Johns Hopkins and Stanford University.
Source: Johns Hopkins University