Gregory M Slabaugh

US Patent:

7333105, Feb 19, 2008

Filed:

Feb 24, 2005

Appl. No.:

11/067151

Inventors:

Gregory G. Slabaugh - Princeton NJ, US
Gozde Unal - West Windsor NJ, US
Jason Jenn-Kwei Tyan - Princeton NJ, US

Assignee:

Siemens Medical Solutions USA, Inc. - Malvern PA

International Classification:

G06T 17/00
G06K 9/34

US Classification:

345420, 382164

Abstract:

A method for three dimensional image segmentation of a volume of interest includes providing a three dimensional image of the volume of interest, providing an initial polyhedron having a plurality of mesh vertices within the three dimension image and determining an image-based speed at each vertex of the polyhedron using an ordinary differential equation (ODE) that describes the vertex motion of the polyhedron. The method further includes determining a regularization term at each vertex of the polyhedron, updating the plurality of mesh vertices of the polyhedron, integrating the image-based speed of each vertex over a face of the polyhedron, and determining an output polyhedron approximating a shape of the volume of interest.

US Patent:

7400767, Jul 15, 2008

Filed:

Jul 12, 2006

Appl. No.:

11/456947

Inventors:

Gregory G. Slabaugh - Princeton NJ, US
Gozde Unal - West Windsor NJ, US

Assignee:

Siemens Medical Solutions USA, Inc. - Malvern PA

International Classification:

G06K 9/34

US Classification:

382173, 382203

Abstract:

A system and method for graph cut image segmentation using a shape prior is provided. In this method, an initial shape is applied to a portion of an image to be segmented. A narrowband is formed around a border of the shape, and a minimized graph cut is calculated for a portion of the image within the narrowband. The shape is then adjusted using the shape prior to fit the minimized graph cut. This method can be iteratively performed so that the shape evolves to segment an object from an image. The shape prior can be a parametric shape, such as an ellipse, or a statistical shape eigenspace calculated based on one or more training shapes.

US Patent:

7620226, Nov 17, 2009

Filed:

Oct 25, 2005

Appl. No.:

11/258422

Inventors:

Gozde Unal - West Windsor NJ, US
Gregory G. Slabaugh - Princeton NJ, US
Hong Shen - Plainsboro NJ, US

Assignee:

Siemens Medical Solutions USA, Inc. - Malvern PA

International Classification:

G06K 9/00

US Classification:

382128

Abstract:

An exemplary method of detecting a structure in a three-dimensional (3D) computed tomography (“CT”) knee image is provided. An intensity prior mask is created based on a two-dimensional (2D) slice of the CT image. A seed point in the 2D slice is received or estimated. The seed point is in a Hoffa's pad region. A geometric prior mask is created based on the seed point. The intensity prior mask and the geometric mask are combined to form a region mask. A level-set contour is segmented from the 2D slice. The step of segmenting is constrained by the region mask.

US Patent:

7623127, Nov 24, 2009

Filed:

Aug 22, 2006

Appl. No.:

11/466211

Inventors:

Hui Xie - Plainsboro NJ, US
Jin Zhou - Forest Hills NY, US
Gregory G. Slabaugh - Princeton NJ, US
Gozde Unal - West Windsor NJ, US
Tong Fang - Morganville NJ, US

Assignee:

Siemens Medical Solutions USA, Inc. - Malvern PA

International Classification:

G06T 15/00
G06T 17/00
G09G 5/00

US Classification:

345419, 345420, 345428, 345581

Abstract:

A method and apparatus for rounding a sharp edge of a model of an object is disclosed whereby a ball is propagated in a desired direction along the edge to be smoothed. The position of the ball at each point in its propagation is noted and, as a result, a virtual tunnel through which the ball passed may be constructed. Points on the sides of the surface of the object in proximity to the sharp edge are then projected onto the virtual tunnel by connecting with a line each point in proximity to the sharp edge to the center of the tunnel. New projected points are created at each position on the surface of the tunnel where the lines intersect that surface. The original points along the sharp edge are then hidden or deleted and the points along the virtual tunnel are connected via well-known surface reconstruction methods. In this way, a sharp edge of a triangle mesh model are advantageously smoothed.

US Patent:

7711148, May 4, 2010

Filed:

Dec 4, 2006

Appl. No.:

11/566363

Inventors:

Gregory G. Slabaugh - Princeton NJ, US
Koon Yin Kong - Atlanta GA, US
Gozde Unal - West Windsor NJ, US
Tong Fang - Morganville NJ, US

Assignee:

Siemens Medical Solutions USA, Inc. - Malvern PA

International Classification:

G06K 9/00
H05G 1/64

US Classification:

382103, 382128, 378 983

Abstract:

A method of tracking a guidewire in video imagery includes: obtaining a first video image including pixels associated with features of a guidewire; selecting a set of parameters to define an open curve on the first video image; determining a feature map of the first video image using phase congruency; and updating the parameters of the open curve using the feature map to align the open curve to the pixels associated with the features of the guidewire.

US Patent:

7720268, May 18, 2010

Filed:

Jun 30, 2006

Appl. No.:

11/427960

Inventors:

Gregory G. Slabaugh - Princeton NJ, US
Gozde Unal - West Windsor NJ, US
Tong Fang - Morganville NJ, US
Michael Wels - Itzgrund, DE

Assignee:

Siemens Corporation - Iselin NJ

International Classification:

G06K 9/00

US Classification:

382128

Abstract:

A method for segmenting a digitized ultrasound image includes providing a digitized in-phase/quadrature ultrasound image comprising a plurality of intensities defined on an N-dimensional grid, decorrelating the ultrasound image wherein spatial correlations are substantially reduced in the intensity data, modeling the decorrelated image intensities with a statistical distribution, propagating, an active contour in the image where the contour segments the image, where the contour is propagated based on the statistical distributions of intensity data inside and outside the active contour.

US Patent:

7787678, Aug 31, 2010

Filed:

Oct 5, 2006

Appl. No.:

11/538921

Inventors:

Gozde Unal - West Windsor NJ, US
Gregory G. Slabaugh - Princeton NJ, US
Tong Fang - Morganville NJ, US

Assignee:

Siemens Corporation - Iselin NJ

International Classification:

G06K 9/00
G06F 7/60

US Classification:

382128, 703 2

Abstract:

Certain exemplary embodiments can comprise a method, which can comprise, automatically determining a target object from an image via an application of an iterative minimization algorithm to an energy function. Each iteration of the iterative minimization algorithm can comprise a comparison of a model with an estimated contour of the target object.

US Patent:

7801343, Sep 21, 2010

Filed:

Nov 2, 2006

Appl. No.:

11/555806

Inventors:

Gozde Unal - West Windsor NJ, US
Yan Yang - Atlanta GA, US
Gregory G. Slabaugh - Princeton NJ, US
Tong Fang - Morganville NJ, US

Assignee:

Siemens Medical Solutions USA, Inc. - Malvern PA

International Classification:

G06K 9/00
A61B 5/05

US Classification:

382128, 600407

Abstract:

A method and apparatus for automatically detecting stent struts in an image is disclosed whereby the inner boundary, or lumen, of an artery wall is first detected automatically and intensity profiles along rays in the image are determined. In one embodiment, detection of the lumen boundary may be accomplished, for example, by evolving a geometric shape, such as an ellipse, using a region-based algorithm technique, a geodesic boundary-based algorithm technique or a combination of the two techniques. Once the lumen boundary has been determined, in another embodiment, the stent struts are detected using a ray shooting algorithm whereby a ray is projected outward in the OCT image starting from the position in the image of the OCT sensor. The intensities of the pixels along the ray are used to detect the presence of a stent strut in the image.