Other challenges include providing computers with the ability to understand the context in which gestures are made and to discern intended versus unintended ones.
You don't want the robot to hand the surgeon a hemostat when she starts talking to another person in the operating room and makes "conversational gestures," Wachs says.
So, will robots replace scrub nurses? Maybe, he says. It will be difficult for a robot to replace a nurse who has extensive experienced with a particular surgeon, he acknowledges. However, many scrub nurses have only limited experience with a given surgeon, which increases the likelihood of misunderstandings, delays, and errors. In those cases, he says, "a robotic scrub nurse could be better."
He's already developed a prototype, and he tells HealthLeaders Media that he expects to have a fully operational one the OR in four to five years.
Robotic scrub nurses aren't a new concept, but most research has focused on voice recognition; little has been done in gesture recognition, Wachs explains. "Another big difference between our focus and the others' is that we are also working on prediction, to anticipate what images the surgeon will need to see next and what instruments will be needed,"Wachs says.
So what sorts of surgery will be most appropriate for this technology?
He points out that, to date, robotics have been deployed primarily for endoscopic and laparoscopic surgeries, as has the Da Vinci. "Open surgeries, such as trauma surgeries, have not used robots in any way. This is the main objective of our system," he tells HealthLeaders Media. "In general, I believe that there is no limit to the extent that robots can be used in the OR. Are they going to replace surgeons? Most likely they are going to complement them."