For the first time ever, a Robot Outperformed a Surgeon in a Precision Training Task.
During surgery, surgeons typically have to make relatively large incisions, whereas the miniature instruments of a robot can fit through smaller incisions. In addition, given the benefits of robotic systems, surgeons are increasingly using remote-controlled robotic arms to perform surgery, combining the precision of an experienced human with the minimal invasiveness of a small robotic arm. Nonetheless, in these cases, the surgeon controls the robot, and a fully automated mechanical system capable of outperforming surgeons in terms of precision has yet to be realized.
However, a recent breakthrough suggests that robots may soon outperform humans. A multinational team of researchers reported the results of a study in which a robot was able to complete a common surgery training task with the same accuracy as an experienced surgeon while completing the task more quickly and consistently.
In their study, the researchers used a standard da Vinci robotic system, a common model used for robot-assisted surgery, and strategically placed 3-D printed markers on its robotic arm. Using a color and depth sensor, the team was able to track its movements. They then used a machine-learning algorithm to analyze the arm's movements. According to the results, the trained model can reduce the mean tracking error by 78 percent, from 2.96 millimeters to 0.65 millimeters.
The researchers then tested their system against an experienced surgeon who had performed over 900 surgeries and nine volunteers with no surgical experience. Participants in the study were asked to complete a peg transfer task, which is a standardized test for training surgeons that involves moving six triangular blocks from one side of a pegboard to the other and back again. While the task appears simple enough, it necessitates millimeter accuracy.
In terms of speed, the surgeon outperformed the automated robot using only one arm. However, the robot outperformed the surgeon in more complex tasks involving two arms.
As far as the researchers know, this is the first time a robot has outperformed a human in a surgery-related training task.