Researchers at Johns Hopkins University have taken a significant step forward in the field of surgical robotics. They have developed an AI-powered robot capable of performing an entire surgical procedure independently, achieving a perfect success rate in multiple tests on realistic models.
The robot, known as SRT-H (Surgical Robot Transformer-Hierarchy), learned how to carry out complex operations by analyzing hours of surgical videos. It can identify vital structures such as arteries and ducts, manipulate tissues with precision, and execute all necessary steps.
Notably, it performed the full procedure guided only by voice commands, similar to instructions from a surgical team.
What’s also promising is the robot’s ability to understand and adapt during surgery. Unlike earlier systems made for specific tasks, SRT-H can grasp the overall process and make real-time adjustments when tissues appear different from expectations. This flexibility is essential for applying such systems in real-world clinical environments.
Built on sophisticated machine learning architectures similar to those powering conversational AI like ChatGPT, SRT-H can respond to spoken commands and learn during the procedure. In tests, it achieved a 100% success rate when performing the same operation on pig organs—models that closely resemble human tissues.
While it currently takes longer than a human surgeon to complete procedures, the accuracy and consistency of its work are comparable to experienced professionals. It’s predicted that within the next decade, autonomous surgical robots could perform a variety of procedures with minimal supervision, potentially transforming medical technology and expanding access to specialized surgical care worldwide.
Learn more on the John Hopkins University Hub official website here.
Leave a Reply