How flight simulation tech turns robots into surgeons

How flight simulation tech turns robots into surgeons

Borrowing techniques from flight simulation technology, CITRIS researchers have now developed an undulating platform to train surgical robots to operate on breathing, pulsing patients.

Wired, March 1, 2018 – THE ROBOTIC PLATFORM heaves, as if breathing. Atop it stretches a piece of white gauze with a blue line painted down the middle. Along this line another robot snips with little surgical scissors, waiting for the platform to come to a brief rest before making a cut. And another snip, and rest. And another, and rest, on down the line.

This could be you one day. Not that you’ll turn into a robot—you may go under the knife of a machine working as a surgical assistant. And it’s machines like this heaving robotic platform—actually adapted from the technology that powers flight simulators—that you’ll have to thank.

A subtle challenge of operating on humans is that their lungs keep breathing and their hearts keep beating—that is, if the surgeon is doing their job right. When the chest heaves or blood pumps, the surgeon has to compensate for that movement.

This new robot mimics that movement. It’s a kind of a Stewart platform, a normally hefty pneumatic device that powers things like immersive flight simulators. But for this study, the researchers took the concept and shrunk it down to a 6-inch-wide device, opting for servo motors instead of pneumatic power. The machine costs just $250.

Fake patient in hand, the researchers first had a human practice cutting on it in a straight line with the da Vinci surgery robot. “We had this nice sinusoidal motion of this platform, something mimicking a heartbeat, and we expected him to follow the motion,” says UC Berkeley computer scientist Sanjay Krishnan. “Instead he would wait for the platform to return to a certain position and then do something really quickly, and then wait again.” Meaning, instead of constantly cutting, the surgeon waited for the machine to reach a lull, then cut.

The researchers took the data from watching the surgeon’s movements and developed algorithms that could mimic his strategy for cutting along a line. They also developed algorithms for cutting continuously, without waiting for lulls. “If you were to design an algorithm to exactly track the motion, first it fails a couple times where it outright misses the line,” says Krishnan. This method of continuously cutting is also more than twice as inaccurate as what the surgeon demonstrated. (The system can also grab grains of rice off the platform, by the way, as you can see above.)

Read the full story on Wired >