Robot-assisted subretinal injection system: development and preliminary verification

Abstract

Abstract Background To design and develop a surgical robot capable of assisting subretinal injection. Methods A remote center of motion (RCM) mechanical design and a master-slave teleoperation were used to develop and manufacture the assisted subretinal surgery robot (RASR). Ten fresh isolated porcine eyes were divided into the Robot Manipulation (RM) group and Manual Manipulation (MM) group (5 eyes for each group), and subretinal injections were performed by the robot and manual manipulation methods, respectively. A preliminary verification of the robot was performed by comparing the advantages and disadvantages of the robot manipulation and manual manipulation by using optical coherent tomography (OCT), fundus photography, and video motion capture analysis after the surgery. Results Both the robot and the manual manipulation were able to perform subretinal injections with a 100% success rate. The OCT results showed that the average subretinal area was 1.548 mm2 and 1.461 mm2 in the RM and MM groups, respectively (P > 0.05). Meanwhile the volume of subretinal fluid obtained using the retinal map mode built in OCT was not statistically different between the RM and MM groups (P > 0.05). By analyzing the surgical video using Kinovea, a motion capture and analysis software, the results suggest that the mean tremor amplitude of the RM group was 0.3681 pixels (x direction), which was significantly reduced compared to 18.8779 pixels (x direction) in the MM group (P < 0.0001). Conclusion Robot-assisted subretinal injection system (RASR) is able to finish subretinal injection surgery with better stability and less fatigue than manual manipulation.