Image processing device and method for detecting line structures in an image data set

• US 9,536,318 B2
• Filed: 12/02/2013
• Issued: 01/03/2017
• Est. Priority Date: 12/03/2012
• Status: Active Grant

• Current Assignee: Koninklijke Philips N.V.
• Original Assignee: Koninklijke Philips N.V.
• Inventors: Wiemker, Rafael, Klinder, Tobias
• Primary Examiner(s): HEIDEMANN, JASON E
• Application Number:
  US14/646,044
• Publication Number:
  US 20150302602A1
• Time in Patent Office: 1,128 Days
• Field of Search: None
• US Class Current: 1/1
• CPC Class Codes: G06T 2200/04 involving 3D image data
• CPC Class Codes: G06T 2207/10081 Computed x-ray tomography [CT]
• CPC Class Codes: G06T 2207/20012 Locally adaptive
• CPC Class Codes: G06T 2207/20016 Hierarchical, coarse-to-fine, multiscale or multiresolution image processing; Pyramid transform
• CPC Class Codes: G06T 2207/20221 Image fusion; Image merging
• CPC Class Codes: G06T 2207/30061 Lung
• CPC Class Codes: G06T 7/12 Edge-based segmentation

Abstract

The present invention relates to an image processing device for detecting line structures in an image data set. The device comprises a model definition unit (12) for defining a line model of a line structure to be detected, said line model comprising a number of voxels, a calculation unit (14) for calculating, per voxel of interest of said image data set, several correlation values of a correlation between said line model and an image area around said voxel of interest, said image area comprising a corresponding number of voxels as said line model, wherein for each of a number of different relative orientations of said line model with respect to said image area a respective correlation value is calculated, and a determining unit (16) for determining, per voxel of interest, the maximum correlation value from said calculated correlation values and the corresponding optimal orientation at which said maximum correlation value is obtained.

Claims

1. An image processing device for detecting anatomical line structures in an image data set, comprising:

a processor that;

defines a line model of an anatomical line structure to be detected, said line model comprising a number of voxels; and

said line model configured to define a two-dimensional rectangular line model having a predetermined length and width;

calculates, per voxel of interest of said image data set, several correlation values of a correlation between said line model and an image area around said voxel of interest, said image area comprising a corresponding number of voxels as said line model, wherein for each of a number of different relative orientations of said line model with respect to said image area a respective correlation value is calculated, and wherein said processor is configured to calculate, per pixel of interest of two-dimensional image data set, several correlation values of a correlation between said two-dimensional rectangular line model and a two-dimensional image area around said pixel of interest;

determines, per voxel of interest, a maximum correlation value from said calculated correlation values, which detects the anatomical line structure, and a corresponding optimal orientation at which said maximum correlation value is obtained; and

identifies within the image data set areas of voxels for which substantially the same optimal orientation has been found, determines, per voxel of interest, a local orientedness value and multiplies, per voxel of interest, said local orientedness value with the maximum correlation value to obtain an amplified maximum correlation value.

2. The image processing device as claimed in claim 1, wherein said processor is further configured to calculate, per voxel of interest of said image data set, several correlation values of a linear correlation between said line model and an image area around said voxel of interest.

3. The image processing device as claimed in claim 2, wherein said processor is further configured to calculate, per voxel of interest of said image data set, several correlation values of said linear correlation between said line model and an image area around said voxel of interest by calculating the ratio of the covariance of intensity values of said image area and said line model over the variance of the intensity values of said image area and variance of the intensity values of said line model.

4. The image processing device as claimed in claim 1, wherein said processor is further configured to define a two-dimensional rectangular line model having a larger extension in a direction along the line structure to be detected than in a direction perpendicular to the line structure to be detected.

5. The image processing device as claimed in claim 1, wherein said processor is further configured to repeat the steps of calculating and determining for different two-dimensional image data sets, in particular different image slices of a three-dimensional image data set.

6. The image processing device as claimed in claim 1, wherein said processor is further configured to define a three-dimensional cuboidal line model having a predetermined length, width and height.

7. The image processing device as claimed in claim 1, wherein said processor is further configured to define a map of voxels or an analytical function as line model.

8. The image processing device as claimed in claim 1, wherein said processor is further configured to initially define a general line model of the line structure to be detected and to use the maximum correlation values and the corresponding optimal orientations obtained by use of said general line model to define a refined line model of the line structure to be detected, and configured to repeat the steps of calculating and determining by use of said refined line model.

9. The image processing device as claimed in claim 1, wherein said processor is further configured to calculate said correlation values only for voxels belonging to a predetermined region of interest and to determine if a voxel belongs to said predetermined region of interest based on a mask image indicating if a voxel belongs to said predetermined region of interest or a probability value of the probability that a voxel belongs to said predetermined region of interest.

10. The image processing device as claimed in claim 1, wherein said processor is further configured to calculate, per voxel of interest of said image data set, several correlation values of a correlation between said line model and an image area around said voxel of interest for different relative orientations of said line model with respect to said image area by rotating said line model and/or said image area around said voxel of interest to change the relative orientation.

11. The image processing device as claimed in claim 1, wherein the processor replaces an original value of a voxel with the obtained maximum correlation value for the voxel.

12. The image processing device as claimed in claim 1, wherein the processor employs a line enhancing filter using multiple hypotheses.

13. An image processing method for detecting anatomical line structures in image data set, comprising:

defining a line model of an anatomical line structure to be detected, said line model comprising a number of voxels; and

said line model defining a two-dimensional rectangular line model having a predetermined length and width;

calculating, per voxel of interest of said image data set, several correlation values of a correlation between said line model and an image area around said voxel of interest, said image area comprising a corresponding number of voxels as said line model, wherein for each of a number of different relative orientations of said line model with respect to said image area a respective correlation value is calculated, and wherein per pixel of interest of two-dimensional image data set, several correlation values of a correlation between said two-dimensional rectangular line model and a two-dimensional image area around said pixel of interest are calculated;

determining, per voxel of interest, a maximum correlation value from said calculated correlation values, which detects the anatomical line structure, and a corresponding optimal orientation at which said maximum correlation value is obtained;

determining, per voxel of interest, a local orientedness value;

constructing a new image using as voxel values the orientedness value and multiplying, per voxel of interest, said local orientedness value with the obtained maximum correlation value.

14. A non-transitory computer readable storage medium storing a computer program for causing a computer to carry out the steps of:

defining a line model of an anatomical line structure to be detected, said line model comprising a number of voxels; and

said line model defining a two-dimensional rectangular line model having a predetermined length and width;

calculating, per voxel of interest of said image data set, several correlation values of a correlation between said line model and an image area around said voxel of interest, said image area comprising a corresponding number of voxels as said line model, wherein for each of a number of different relative orientations of said line model with respect to said image area a respective correlation value is calculated, and wherein per pixel of interest of two-dimensional image data set, several correlation values of a correlation between said two-dimensional rectangular line model and a two-dimensional image area around said pixel of interest are calculated;

determining, per voxel of interest, a maximum correlation value from said calculated correlation values, which detects the anatomical line structure, and a corresponding optimal orientation at which said maximum correlation value is obtained;

determining, per voxel of interest, a local orientedness value;

constructing a new image using as voxel values the orientedness value and multiplying, per voxel of interest, said local orientedness value with the obtained maximum correlation value.

15. The non-transitory computer readable storage medium as claimed in claim 14, further comprising:

performing the calculating and determining steps with a non-Hessian filter.

16. The non-transitory computer readable storage medium as claimed in claim 14, further comprising:

calculating, for each voxel, a line measure by correlating a model template of the line structure with observed image data.