Doctor of Philosophy
Dr. Terry Peters, Dr. Roy Eagleson
Surgical image visualization and interaction systems can dramatically affect the efficacy and efficiency of surgical training, planning, and interventions. This is even more profound in the case of minimally-invasive surgery where restricted access to the operative field in conjunction with limited field of view necessitate a visualization medium to provide patient-specific information at any given moment. Unfortunately, little research has been devoted to studying human factors associated with medical image displays and the need for a robust, intuitive visualization and interaction interfaces has remained largely unfulfilled to this day. Failure to engineer efficient medical solutions and design intuitive visualization interfaces is argued to be one of the major barriers to the meaningful transfer of innovative technology to the operating room. This thesis was, therefore, motivated by the need to study various cognitive and perceptual aspects of human factors in surgical image visualization systems, to increase the efficiency and effectiveness of medical interfaces, and ultimately to improve patient outcomes. To this end, we chose four different minimally-invasive interventions in the realm of surgical training, planning, training for planning, and navigation: The first chapter involves the use of stereoendoscopes to reduce morbidity in endoscopic third ventriculostomy. The results of this study suggest that, compared with conventional endoscopes, the detection of the basilar artery on the surface of the third ventricle can be facilitated with the use of stereoendoscopes, increasing the safety of targeting in third ventriculostomy procedures. In the second chapter, a contour enhancement technique is described to improve preoperative planning of arteriovenous malformation interventions. The proposed method, particularly when combined with stereopsis, is shown to increase the speed and accuracy of understanding the spatial relationship between vascular structures. In the third chapter, an augmented-reality system is proposed to facilitate the training of planning brain tumour resection. The results of our user study indicate that the proposed system improves subjects' performance, particularly novices', in formulating the optimal point of entry and surgical path independent of the sensorimotor tasks performed. In the last chapter, the role of fully-immersive simulation environments on the surgeons' non-technical skills to perform vertebroplasty procedure is investigated. Our results suggest that while training surgeons may increase their technical skills, the introduction of crisis scenarios significantly disturbs the performance, emphasizing the need of realistic simulation environments as part of training curriculum.
Abhari, Kamyar, "Visual Perception and Cognition in Image-Guided Intervention" (2014). Electronic Thesis and Dissertation Repository. 2432.