Type of Culminating Activity

Graduate Student Project

Graduation Date

5-8-2012

Degree Title

Master of Science in Computer Science

Department

Electrical Engineering

Major Advisor

Elisa Barney Smith

Abstract

Biomedical researchers constantly search for new methods to diagnose the extent of joint injuries in live human subjects. In order to achieve this, the researchers need to know the accurate three dimensional kinematic data of bones and joints and to accurately quantify how bones in a joint move relative to one another during dynamic activities. Algorithms have been developed previously to estimate the exact spatial position and movement of the bones. These methods involved generation of digitally reconstructed radiographs (DRR) from a three dimensional CT scan image of human joint with known position and orientation and comparing it with the original two dimensional video fluoroscopy (video X-ray) image. This way, a DRR image could be found which looks similar to the fluoroscopy frame of the knee joint and thus the position and orientation of the bone could be discovered.

Previous research has been done for finding the dimensions and the orientation of the bone by making use of a sequential Monte Carlo method [1] and Swarm Intelligence techniques in a parallel computing environment [2]. These methods are a little hard to understand by a person not involved in image processing. This project is about developing a user friendly GUI to help people understand the scope of the problem on which researchers are working. The GUI displays the bones in both three dimensional and two dimensional space, allowing the human user to do the search which was previously done automatically by the computer using image feature information. The user of the GUI will estimate the position of a given bone, given images of a bone in 3D and 2D space viewed at perspectives chosen by the user.

Download

Included in

Biomedical Commons

Share

COinS