Abstract:
This paper introduces a workflow for robot assisted neurosurgery. The system consists of a robot for precise positioning and a vision system for intro-operative registration. The vision system uses two cameras and templates to automatically track dynamic targets. In order to accurately position the end of the robot during operation, the robot arm must occupy the least space, be easy to move and be well balanced. Cylindrical configuration is chosen in this case as a suitable type of arm. Based on the Jacobian matrix of robot and the conditioning number of the matrix, the optimal dexterous region of the robot is found to satisfy desired dexterity. The successful clinical application verifies the effectiveness of the robot system