record – LABS

Thesis Info

Thesis Supervisor: Dr. David Steinman, Professor, Mechanical and Industrial Engineering, University of Toronto

Other Supervisor(s): Dr. Peter Coppin, Assistant Professor, Faculty of Design, OCAD University and Professor (status appointment), Mechanical and Industrial Engineering, University of Toronto

Keywords: computational fluid dynamics; feature-tracking; hemodynamics; illustration-inspired; visualization

Abstract: 200-500 words: Image-based computational fluid dynamics is an invaluable tool in the study and evaluation of hemodynamic factors involved in cardiovascular disease. Simulating blood flow inside a patient-specific model provides an understanding of the blood's complex behaviour without resorting to risky invasive procedures. However, detailed hemodynamic data can be difficult to present clearly using conventional flow visualization techniques. Many approaches use single snapshots in time which fail to capture the time-varying qualities of the flow dynamics; and animations retain temporal information but are constrained to a single point of view. To address these concerns, we took cues from illustration and comics to explore data-driven paradigms for visualizing blood flow. We programmatically reproduced illustrative techniques using modern computer graphics capabilities and computer vision-inspired feature-tracking algorithms. The culmination of this work produced an interactive web-based tool that allows users to explore the hemodynamics of cerebral aneurysms in a unique, immediate, and novel way.