The basic premise of Magnetic Resonance Imaging (MRI) - guided neurosurgery is that the location of a surgical instrument can be shown on an image display monitor relative to a detailed depiction of the internal cranium. Precision and potential for tele-surgery are the prime motivations for applying robots in the MRI environments. Typically neurosurgeries are performed in open-bore MRI scanners. This results in the use of preoperative MRI images during the procedures. Use of closed-bore scanners would eliminate this concern;​ however there is no space for the neurosurgeon to perform operations. Therefore remote-control surgery would be the appropriate method to be used in closed-bore MRI-based surgery.

In this dissertation the design and control paradigms of a novel modular tele-robotic system for closed-bore MRI-guided neurosurgery are presented. Candidate neurosurgical procedures enabled by this system would include thermal ablation, radiofrequency ablation, deep brain stimulators, and targeted drug delivery. A new infrastructure for MRI-guided intervention is also developed to address clinical requirements in a typical closed-bore MRI environment.

The design paradigm is fundamentally based on a modular design configuration for the slave manipulator performing the required task inside MR scanner. Navigation and operating modules were designed to undertake the alignment and advancement of the surgical needle respectively.

The control paradigm was developed based on two novel control methods including fiducials tracking and semi-autonomous motion. In the former, the surgeon could manipulate the needle inside the MRI scanner while relative position of the needle and the target are visualized on a display. In the latter, the needle is manipulated

autonomously based on feedback from MR images to the controller. Two MR-compatible actuation systems that include ultrasonic motors and hydraulic/pneumatic cylinders were developed. A series of the experimental tests were conducted to evaluate MR-compatibility of an ultrasonic motor. The results show that the actuation of the motor only slightly deteriorated the MR image, and no image shift and significant degradation of signal-to-noise ratio was observed.

This research provides a first complete surgical system for closed-bore MRI-based neurosurgery. The main contributions are the system MR-compatibility and the modular design and control paradigms.