Detail preserving anatomical markers in a virtual reality nasendoscopy simulation

摘要

Nasendoscopy is commonly used to examine the sinus cavity. The anatomy of the sinuses is highly varied and complex. It consists of intricate, interconnected structures and anatomical landmarks that can be difficult to identify; firstly because of the complexity, and secondly because of the diversity across patients. Consequently, teaching medical trainees to recognize these anatomical landmarks requires a way of highlighting landmarks without concealing important, and often subtle, details. Further, when using virtual reality simulations to assess a trainee's knowledge of the anatomy, it is important to annotate 3-dimensional models in a manner that does not reveal a target location prematurely. Here we describe a simple new method that uses recent advances in the programmability of the graphics processing unit, to efficiently, clearly and accurately display visual markers. Moreover, our method allows markers to be positioned simply and intuitively using a pair of haptic devices without requiring complex modifications to surface textures.