Date of Award

Spring 2012

Document Type


Degree Name

Doctor of Philosophy (PhD)


Electrical & Computer Engineering


Computer Engineering

Committee Director

Frederic D. McKenzie

Committee Member

Jen-Kuang Huang

Committee Member

Yuzhong Shen

Committee Member

Jiang Li


Dental implantation is now recognized as the standard of the care for tooth replacement. Although many studies show high short term survival rates greater than 95%, long term studies (> 5 years) have shown success rates as low as 41.9%. Reasons affecting the long term success rates might include surgical factors such as limited accuracy of implant placement, lack of spacing controls, and overheating during the placement.

In this dissertation, a comprehensive solution for improving the outcome of current dental implantation is presented, which includes computer-aided preoperative planning for better visualization of patient-specific information and automated robotic site-preparation for superior placement and orientation accuracy. Surgical planning is generated using patient-specific three-dimensional (3D) models which are reconstructed from Cone-beam CT images. An innovative image-guided robotic site-preparation system for implants insertion is designed and implemented. The preoperative plan of the implant insertion is transferred into intra-operative operations of the robot using a two-step registration procedure with the help of a Coordinate Measurement Machine (CMM). The natural-root implants mimic the root structure of natural teeth and were proved by Finite Element Method (FEM) to provide superior stress distribution than current cylinder-shape implants. However, due to their complicated geometry, manual site-preparation for these implants cannot be accomplished. Our innovative image-guided robotic implantation system provides the possibility of using this advanced type of implant.

Phantom experiments with patient-specific jaw models were performed to evaluate the accuracy of positioning and orientation. Fiducial Registration Error (FRE) values less than 0.20 mm and final Target Registration Error (TRE) values after the two-step registration of 0.36±0.13 mm (N=5) were achieved. Orientation error was 1.99±1.27° (N=14). Robotic milling of the natural-root implant shape with single- and double-root was also tested, and the results proved that their complicated volumes can be removed as designed by the robot. The milling time for single- and double-root shape was 177 s and 1522 s, respectively.


In Copyright. URI: This Item is protected by copyright and/or related rights. You are free to use this Item in any way that is permitted by the copyright and related rights legislation that applies to your use. For other uses you need to obtain permission from the rights-holder(s).