Image analysis – Applications – Biomedical applications
Reexamination Certificate
2001-04-09
2004-02-10
Patel, Jayanti K. (Department: 2625)
Image analysis
Applications
Biomedical applications
C600S407000
Reexamination Certificate
active
06690816
ABSTRACT:
FIELD OF THE INVENTION
The present invention is directed generally to image processing systems and methods, and more particularly, to systems and methods for processing images to represent and analyze at least one tubular-shaped object.
BACKGROUND OF THE INVENTION
Tubular-shaped objects, herein referred to as tubular objects, may be found in image data in a variety of imaging applications. For example, tubular objects may be commonly found in medical images. Examples of tubular objects in medical images may include vessels, bronchi, bowels, ducts, nerves and specific bones. Representation and analysis of tubular objects in medical images can aid medical personnel in understanding the complex anatomy of a patient and facilitate medical treatments. Remote sensing is another example application. Images of terrain containing natural features, for example, rivers and/or streams, and/or man-made features, such as tunnels and/or roads, may also be considered tubular objects. Representation and analysis of tubular objects in the remote sensing application may be useful in the creation of digital maps. Tubular objects may exist in many other diverse imaging applications, such as, for example, integrated circuit manufacturing, electron microscopy, etc. Processing of tubular objects in any of these imaging applications may provide information to maximize the utility of these images.
Known approaches for generating representations of objects in images include techniques characterized as “model-based” and “model-free.” Model-based methods can be difficult to apply to most tubular object representation and analysis tasks since the global arrangement of most tubular objects varies significantly from image to image. For example, in medical imaging, the coiling of the bowel or a kidney's vascular tree varies significantly from patient to patient. There typically is not a one-to-one correspondence between the arrangement of most vessels among different patients.
Model-free techniques include thresholding, region growing, morphological, and image filtering techniques. Global threshold techniques can lack feasibility for extracting vessels from Magnetic Resonance Angiograms (MRAs) due to large-scale intensity variations which may occur in that imaging modality. Region growing methods appear to have difficulties with small vessels, especially for “one-voxel” vessels whose diameter is approximately the resolution of the imaging device. Additionally, thresholding and region growing methods may sometimes not be stable and may not form representations of tubes. Morphological and image filtering techniques may assume that a fixed, basic shape has been extruded along a path to generate an object. Such methods can be over-constrained and present a difficulty in handling a wide range of tube sizes and/or variations in cross-sectional intensities. While these methods can form symbolic representations, the stability and utility of the symbolic representations remains to be developed.
Available techniques for tubular object representation may be unable to form stable representations in the presence of noise, be computationally inefficient, unable to provide three-dimensional connectivity information, exploit the geometry of tubes and scale invariance, or operate independently of the data source.
In view of the foregoing, there is a need for an improved system and method for stable, accurate, and fast representation and analysis of tubular objects in multi-dimensional images.
SUMMARY
The present invention provides a system and method for processing a multi-dimensional image containing at least one tubular object.
As embodied and broadly described herein, certain aspects of the invention are directed to a system which processes at least one tubular object found in a multi-dimensional image.
In one aspect of the invention, a method for processing at least one tubular object in a multi-dimensional image is presented which includes establishing a seed point in a multi-dimensional image, searching for an extremum point corresponding to a tubular object having a central curve of intensity extrema, and extracting: 1) a one-dimensional central track corresponding to the tubular object, and 2) extents corresponding to a plurality of cross-sections along the tubular object, where the plurality of cross-sections intersects the central track.
In another aspect of the invention, a method for processing at least one tubular objects in a multi-dimensional image is presented which includes establishing a seed point in a multi-dimensional image, searching for an extremum point corresponding to a tubular object having a central curve of intensity extrema, extracting: 1) a one-dimensional central track corresponding to the tubular object, and 2) extents corresponding to a plurality of cross-sections along the tubular object, where the plurality of cross-sections intersects the central track, generating symbolic representations of the tubular object, and optionally performing at least one of a set operation, a numeric operation, and a graph operation on the symbolic representations.
In another aspect of the invention, a system for processing at least one tubular object in a multi-dimensional image is presented which includes a computer processor, a memory functionally coupled to the computer processor, wherein the memory stores a multi-dimensional image and instructions to be executed by the computer processor, for establishing a seed point in the multi-dimensional image, searching for an extremum point corresponding to a tubular object having a central curve of intensity extrema, and extracting: a) a one-dimensional central track corresponding to the tubular object, and b) extents corresponding to a plurality of cross-sections along the tubular object, where the plurality of cross-sections intersects the one-dimensional central track.
In another aspect of the invention, a system for processing at least one tubular object in a multi-dimensional image is presented which includes a computer processor, a memory functionally coupled to the computer processor, wherein the memory stores a multi-dimensional image and instructions to be executed by the computer processor, for establishing a seed point in the multi-dimensional image, searching for an extremum point corresponding to a tubular object having a central curve of intensity extrema, extracting: a) a one-dimensional central track corresponding to the tubular object, and b) extents corresponding to a plurality of cross-sections along the tubular object, where the plurality of cross-sections intersects the one-dimensional central track, generating symbolic representations of the tubular object, and optionally performing at least one of a set operation, a numeric operation, and a graph operation on the symbolic representations.
It is to be understood that both the foregoing general description and the following detailed description are exemplary and explanatory only and are not restrictive of the invention.
REFERENCES:
patent: 5768405 (1998-06-01), Makram-Ebeid
patent: 5891030 (1999-04-01), Johnson et al.
patent: 6169917 (2001-01-01), Masotti et al.
patent: 6212420 (2001-04-01), Wang et al.
patent: 6246784 (2001-06-01), Summers et al.
patent: 6381586 (2002-04-01), Glasserman et al.
patent: WO 98/37517 (1998-08-01), None
patent: WO 00/55814 (2000-09-01), None
Aylward Stephen R.
Bullitt Elizabeth
Fritsch Daniel
Pizer Stephen M.
Jenkins & Wilson & Taylor, P.A.
Patel Jayanti K.
Tabatabai Abolfazl
The University of North Carolina at Chapel Hill
LandOfFree
Systems and methods for tubular object 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 Systems and methods for tubular object processing, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Systems and methods for tubular object processing will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-3297554