Paper
12 March 2008 Image-based path planning for automated virtual colonoscopy navigation
Author Affiliations +
Abstract
Virtual colonoscopy (VC) is a noninvasive method for colonic polyp screening, by reconstructing three-dimensional models of the colon using computerized tomography (CT). In virtual colonoscopy fly-through navigation, it is crucial to generate an optimal camera path for efficient clinical examination. In conventional methods, the centerline of the colon lumen is usually used as the camera path. In order to extract colon centerline, some time consuming pre-processing algorithms must be performed before the fly-through navigation, such as colon segmentation, distance transformation, or topological thinning. In this paper, we present an efficient image-based path planning algorithm for automated virtual colonoscopy fly-through navigation without the requirement of any pre-processing. Our algorithm only needs the physician to provide a seed point as the starting camera position using 2D axial CT images. A wide angle fisheye camera model is used to generate a depth image from the current camera position. Two types of navigational landmarks, safe regions and target regions are extracted from the depth images. Camera position and its corresponding view direction are then determined using these landmarks. The experimental results show that the generated paths are accurate and increase the user comfort during the fly-through navigation. Moreover, because of the efficiency of our path planning algorithm and rendering algorithm, our VC fly-through navigation system can still guarantee 30 FPS.
© (2008) COPYRIGHT Society of Photo-Optical Instrumentation Engineers (SPIE). Downloading of the abstract is permitted for personal use only.
Wei Hong "Image-based path planning for automated virtual colonoscopy navigation", Proc. SPIE 6916, Medical Imaging 2008: Physiology, Function, and Structure from Medical Images, 69160F (12 March 2008); https://doi.org/10.1117/12.769841
Advertisement
Advertisement
RIGHTS & PERMISSIONS
Get copyright permission  Get copyright permission on Copyright Marketplace
KEYWORDS
Cameras

Image segmentation

Virtual colonoscopy

Colon

Endoscopy

Image processing algorithms and systems

Detection and tracking algorithms

RELATED CONTENT


Back to Top