Computer graphics processing and selective visual display system – Computer graphics processing – Three-dimension
Reexamination Certificate
2000-01-04
2002-12-17
Vo, Cliff N. (Department: 2671)
Computer graphics processing and selective visual display system
Computer graphics processing
Three-dimension
Reexamination Certificate
active
06496188
ABSTRACT:
BACKGROUND OF THE INVENTION
The invention relates to an image processing method for processing an image representing a tubular structure and for constructing a path related to said structure. The invention also relates to an image processing system to perform said method. The invention further relates to medical CT or MR apparatus having processing means for constructing 3D-images and processing means for performing virtual endoscopy based on said method.
The invention finds its application in the industry of medical apparatus.
An image processing method for computation of a flight path in medical images for virtual endoscopy is already known from the publication “Automated Flight Path Planning for Virtual Endoscopy” by David S. PAIK, Christopher F. BEAULIEU et alii, in Med. Phys. 25 (5), May 1998, pp.629-637. This publication discloses an automatic computation of a flight path for guiding virtual endoscopic exploration of three-dimensional medical images. This flight path finding method, for positioning a virtual camera flying through images, includes a medial axis transform which provides a first path and an iterative correction of said first path toward the medial axis. More specifically, this method comprises steps of: segmentation of the volume by an operation of region growing and “bubble” removal; computation of a first path by connecting a starting voxel to an end voxel; iterative bringing towards the 3-D medial axis, referred to as improvement of the first path; smoothing the path to determine positions along the medial axis to form said flight path; determination of the virtual camera orientation (direction and twist) along the flight path.
This method first defines a structure of interest by a region growing algorithm which segments the 3-D image starting from seed voxels and which grows regions of interest by connecting to each seed voxel 26 voxels meeting a threshold criterion and by removing inhomogeneities called bubbles; then this method computes an Euclidean distance map in a region of interest and a first Euclidean path that is the shallowest descent in this map; this first Euclidean path is further centralized by an iterative step which removes all surface voxels from the region of interest, determining a new distance map and a new path through this new distance map until the region of interest is thinned away and only the centralized path remains.
This known method provides means for constructing a flight path which is centered at best in said tubular structure. For the application to virtual endoscopy, a centered path is not actually most appropriate to fly the virtual camera through the tubular object because the anatomical tubular object is generally very contorted so that a centered path would be unnecessarily very long and complicated. Besides, certain location found on said centered path would not even be correct to position the virtual camera because for instance said location may be situated just in front of an important protuberance obstructing the view. Moreover, using the computing means known at the present time, the known method needs several minutes to be performed because it necessitates the computation of several distance maps in the iterative step which is proportionally time consuming.
The visualization of volumetric medical image data plays a crucial part in diagnosis operation and therapy planning. The better anatomy and pathology are understood, the more efficient operations can be performed at low risk. Basically, virtual endoscopy gives views of regions of the body dangerous or impossible to reach physically with a camera, such as brain vessels for example, the only requirement being to inject a contrast product in the anatomical objects for better detection. Virtual endoscopy supplies perspective views of human anatomy simulated from the inside of tubular structures. This allows the user to view complex anatomical structures in a comfortable way after data acquisition and almost instantly. A virtual endoscopic system may comport two parts:
Means for an endoscopic path construction providing a continuous location of a flight through the tubular structure of interest;
Three-dimensional interior viewing along the endoscopic path; those views are adjoined creating an animation which simulates a virtual flight through them; the views may be created using 3-D rendering or ray tracing known techniques.
It is an object of the invention to provide a method for acquiring data of a 3-D image representing a contorted tubular structure and for constructing automatically the most appropriate path to fly inside said structure for application to virtual endoscopy in 3-D medical images. It is an other object of the invention to provide said method to be carried out in less time than the known method when using the same kind of computing means.
According to the invention, these objects are achieved by a method in which a first end point and a second end point for a flight path inside the tubular structure are determined and the flight path is determined which is the shortest path between the first and second end points at a maximum, constant distance from the walls of the tubular structure irrespective of the shape of the tubular structure.
An advantage of the method of the invention is that the constructed path in a contorted tubular object shows two important properties. First, this path is maintained at a predetermined distance from the internal walls of the contorted tubular object which may be different from the distance of the exact center line but is more appropriate for obtaining a good visualization of the interior of the object, thereby ensuring that the visualization is not obstructed by a nearby protuberance. Second, this path is also the smallest possible path remaining at the predetermined distance from the internal walls. Another advantage is that this method is completely automated, apart from the need to set only two end conditions, and that only a small part of the 3-D image is visited for finding the path. Another advantage is that only one first computation of a distance map is needed, the first distance map providing the data necessary to perform the further steps of the method. So, the method is not time consuming: only 10 to 30 seconds are necessary to carry out the steps of providing the path with the same kind of computing means as the method of the cited prior art. Another advantage of this method is that it further permits building of a 3-D interior view of the structure along this path without user interactivity other than setting the two end conditions. It also permits visualizing of the inside of the structure without colliding with the internal walls, without crossing the walls and without being obstructed by parts of the contorted tubular object. In particular, this method permits visualizing of the inside of anatomical objects in 3-D CT or MR images in a virtual way and in an automated manner. Thus, this method may be applied to virtual endoscopy.
OBJECTS AND SUMMARY OF THE INVENTION
It is also an object of the invention to provide a system for carrying out this method and to provide a medical apparatus having means for acquiring 3-D images and having a system to process the images according to this method for virtual endoscopy.
This problem is solved respectively by a system as claimed in claim 9, and by an apparatus as claimed in claim 10.
REFERENCES:
patent: WO9613207 (1996-05-01), None
patent: WO9811524 (1998-03-01), None
“Levels-sets Method: Evolving Interfaces in Geometry, Fluid Mechanics, Computer Vision and Material Sciences” by Sethian, Cambridge University Press, 1996b, Chapter 9, pp. 87-95.
“Automatied flight path planning for virtual endoscopy” by Davis S. Paik et al., in Medical Physics, vol. 25(5) May, 1998, pp. 629-637.
“Global Minimum for Active Contour Models: A Minimal Path Approach” by Laurent D. Cohen et al., in International Journal of Computer Vision vol. 24(1) pp. 57-78 (1997).
Cohen Laurent
Deschamps Thomas
Makram-Ebeid Sherif
Koninklijke Philips Electronics , N.V.
Vo Cliff N.
Vodopia John
LandOfFree
Image processing method, system and apparatus for processing... does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Image processing method, system and apparatus for processing..., we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Image processing method, system and apparatus for processing... will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-2927780