Bimanual Robotic Eye Manipulation Using Adaptive Sclera Force Control: Towards Safe Retinal Vein Cannulation

Retinal surgery typically requires bimanual manipulation of tools in the eye. Freehand retinal vein cannulation (RVC) is a highly challenging operation mainly due to typical hand tremors relative to the small size of retinal veins. Robot-assisted technology resolves hand tremor issues and gives ophthalmologists higher positioning resolution to enable RVC. Bimanual robot manipulation of the eyeball typically requires kinematics-based control to maintain each robotic tool's remote center of motion (RCM) constraint and registration between the two robots to avoid scleral injury. Any potential relative movement of the robot base can impact patient safety. To avoid these problems, we developed a bimanual adaptive cooperative (BMAC) control framework. Each robot is independently controlled via a hybrid adaptive position-force control algorithm using fiber Bragg grating-based force-sensing surgical instruments. This algorithm minimizes the tool-sclera interaction forces automatically, resulting in maintaining the sclera forces within a safe threshold and avoiding over-stretch of the sclera, which guarantees patient safety despite the absence of kinematic RCM constraint and registration of the two robots. The effectiveness of this approach is validated through a pilot study with five users in a vessel-following experiment on an eye phantom under a surgical microscope.