Arthroscopic knee surgery can be challenging because there is no intuitive relationship between the arthroscopic image, shown on a screen above the patient, and the camera in the surgeon's hand. As a result, arthroscopic surgeons require extensive training and experience. This thesis describes a computer system to help improve target acquisition in arthroscopy by visualizing the location and alignment of an arthroscope using augmented virtuality. A 3D computer model of the patient's joint (from CT) is shown, along with a model of the tracked arthroscopic probe and the projection of the camera image onto the virtual joint. We performed a user study to determine the effectiveness of this navigated display; the study showed that for novice residents, the navigated display improved target acquisition. However, residents with at least two years of experience performed worse. For surgeons, no effect on performance was found. / Thesis (Master, Computing) -- Queen's University, 2010-11-25 23:29:46.526