Heat diffusion based detection of structures of interest in medical images

• US 7,729,739 B2
• Filed: 11/29/2004
• Issued: 06/01/2010
• Est. Priority Date: 12/03/2003
• Status: Expired due to Fees

• Current Assignee: Board of Trustees of the Leland Stanford Junior University (Stanford University)
• Original Assignee: Board of Trustees of the Leland Stanford Junior University (Stanford University)
• Inventors: Konukoglu, Ender, Paik, David S., Acar, Burak, Beaulieu, Christopher F., Napel, Sandy A.
• Primary Examiner(s): Le; Long V
• Assistant Examiner(s): Gupta; Vani
• Application Number:
  US11/000,515
• Publication Number:
  US 20050149286A1
• Time in Patent Office: 2,010 Days
• Field of Search: 600/407, 600/431, 600/437, 382/128
• US Class Current: 600/407
• CPC Class Codes: G06T 2207/20152 Watershed segmentation
• CPC Class Codes: G06T 2207/30032 Colon polyp
• CPC Class Codes: G06T 7/0012 Biomedical image inspection
• CPC Class Codes: G06T 7/136 involving thresholding
• CPC Class Codes: G06T 7/64 of convexity or concavity
• CPC Class Codes: G06T 7/68 of symmetry

Abstract

A method for detecting and identifying structures of interest such as colonic polyps or similar structures like lung nodules in volumetric (medical) images data is provided. The method includes obtaining a heat diffusion field (HDF) by applying a heat diffusion scheme to a volume of interest that includes structures. The obtained heat diffusion field is then used for identifying a structure of interest from the structures in the volume of interest using a geometrical analysis of the heat diffusion field. The heat diffusion scheme is, at least partly, governed by non-linear diffusion parameters. The identification includes two parts: (i) the computation of a spherical symmetry parameter, and (ii) the performance of a local analysis of the volume of interest and computation of a triangulization parameter.

Claims

1. A computer-implemented method operable on a computer for detecting and identifying structures of interest in computed tomographic (CT) volumetric images, comprising the steps of:

(a) obtaining a heat diffusion field of a volume of interest with structures from a CT volumetric image available on said computer by applying a heat diffusion scheme to said volume of interest with structures;

(b) mapping said obtained heat diffusion field to a vector field of said structures; and

(c) identifying a structure of interest from said structures using a geometrical analysis of said vector field of said structures, wherein said geometrical analysis comprises computing a spherical symmetry parameter from said mapped vector field of said structure of interest, and wherein said identifying comprises performing a local analysis of the area around said structure of interest and computing a triangulization parameter of said area, whereby both said spherical symmetry parameter and said triangulization parameter are used for the identification of said structure of interest.

2. The method as set forth in claim 1, wherein said heat diffusion scheme is at least partly governed by non-linear diffusion parameters.

3. The method as set forth in claim 1, wherein said structures comprises protruding structures and non-protruding structures, and said structure of interest is one of said protruding structures.

4. The method as set forth in claim 3, wherein said protruding structures are polyps, haustral folds or other elongated structures, and wherein said structure of interest is a polyp.

5. The method as set forth in claim 3, wherein said non-protruding structures are flat colon wall regions and regions in between haustral folds.

6. The method as set forth in claim 1, wherein said volume of interest is a colon or a colon wall and said structure of interest is a polyp.

7. The method as set forth in claim 1, wherein said volume of interest is a lung, said structures comprises stand-alone structures, and said structure of interest is a lung nodule.