Automatic image segmentation methods and analysis

• US 9,984,460 B2
• Filed: 09/26/2016
• Issued: 05/29/2018
• Est. Priority Date: 11/23/2007
• Status: Active Grant

• Current Assignee: PME IP Pty., Ltd.
• Original Assignee: PME IP Pty., Ltd.
• Inventors: Westerhoff, Malte, Stalling, Detlev, Seebass, Martin
• Primary Examiner(s): Koval, Melissa
• Assistant Examiner(s): FETZNER, TIFFANY A
• Application Number:
  US15/276,546
• Publication Number:
  US 20170011514A1
• Time in Patent Office: 610 Days
• Field of Search: 324300-322, 600407-435, 382128-131
• CPC Class Codes: G06F 18/24 Classification techniques
• 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 2207/20112 Image segmentation details
• CPC Class Codes: G06T 2207/30008 Bone
• CPC Class Codes: G06T 2207/30101 Blood vessel; Artery; Vein; Vascular
• CPC Class Codes: G06T 7/0012 Biomedical image inspection
• CPC Class Codes: G06T 7/11 Region-based segmentation
• CPC Class Codes: G06V 10/50 by performing operations within image blocks; by using histograms, e.g. histogram of oriented gradients [HoG]; by summing image-intensity values; Projection analysis

Abstract

The invention provides methods and apparatus for image processing that perform image segmentation on data sets in two- and/or three-dimensions so as to resolve structures that have the same or similar grey values (and that would otherwise render with the same or similar intensity values) and that, thereby, facilitate visualization and processing of those data sets.

Claims

1. A non-transitory method of processing anatomical structures present in two dimensional 2D image slices using a digital data processor that permits identification of bone structure, vessel structure, or both bone structure and vessel structure, where bone structure is viewable without vessel structure, where vessel structure is viewable without bone structure, and where the method does not require the anatomical mid sagittal plane to be calculated within any of steps (a)-(f) comprising:

(a) selecting a 2D image slice from a plurality of first 2D image slices spanning a three dimensional (3D) volume, using a digital data processing system configured for image processing;

(b) computing with the digital data processing system a histogram of the selected 2D image slice;

(c) receiving one or more previously determined and identified histograms generated from one or more previously determined second 2D image slices where one or more components are identified in the one or more previously determined second 2D image slices which include one or more of bone, blood vessel structures, vasculature, and blood vessels, where one or more connected components are used in identifying a plurality of pixels in the one or more components, where the plurality of pixels belong to the same one or more components received from the one or more previously determined second 2D image slices;

(d) computing with the digital data processing system a cross correlation between the histogram of step (b) and one or more of the previously determined and identified histograms generated from the one or more second 2D image slices of step (c);

(e) identifying with the digital data processing system, two or more anatomical structures in the 2D image slice of step (b), based on the cross correlation in step (d);

(f) using the digital data processing system in order to;

(1) assign one or more region(s) of the 2D image slice of step (b), corresponding to an anatomical structure of the two or more anatomical structures identified in step (e), to one or more corresponding volumetric regions of the 3D volume;

(2) perform a threshold segmentation on the 2D image slice of step (b);

(3) identify one or more connected components of the anatomical structure identified in step (e) based on the results of the threshold segmentation of step (2); and

(4) display the threshold segmented 2D image slice, where the displayed threshold segmented 2D image slice is capable of producing one of each type of the following 2D image slices;

a bone containing slice, a blood vessel structure containing slice, a vasculature containing slice and a slice depicting blood vessels removed from the displayed threshold segmented 2D image slice based on the one or more connected components identified in step (3).

2. The method of claim 1, where the one or more connected components include both bone structures and vessel structures.

3. The method of claim 2, where the bone structures are removed, from the displayed threshold segmented 2D image slice provided in step (4) by the digital data processing system.

4. The method of claim 3, where only the vessel structures are shown in the displayed threshold segmented 2D image slice provided from step (4) by the digital data processing system.

5. The method of claim 3, where the bone structures are automatically removed from the displayed threshold segmented 2D image slice provided from step (4) by the digital data processing system.

6. The method of claim 3, where the plurality of first 2D images are received from one or more previously performed computed tomography (CT) 3D run-off studies.

7. The method of claim 1, where step (d) involves computing with the digital data processing system a cross correlation between a region in the histogram of the 2D image slice of step (b) and a corresponding region in one or more of the previously determined and identified histograms from the plurality of first 2D image slices that is then displayed to a user of the digital data processing system.

8. The method of claim 7, further comprising checking the assigning and identifying of step (f) with the 3D volume to insure that the assignment(s) and identification(s) are consistent with the imaged anatomy.

9. A system configured for the non-transitory processing of anatomical structures present in two dimensional (2D) anatomical image slices using a digital data processor that permits the identification of bone, soft tissue, or both bone structure and soft tissue, where bone structures are viewable without soft tissue, where soft tissue is viewable without bone structures, and where the method does not require the anatomical mid sagittal plane to be calculated within any of steps (a)-(d) below, the system configured for the processing of anatomical structures comprising:

(a) one or more dynamic memory components;

(b) at least one digital data processor including one or more graphics processing units;

(c) an anatomical image source; and

(d) a non-transitory computer readable program, that when executed on the at least one digital data processor, the non-transitory computer readable program causes the digital data processor to perform each of the method steps (1) through (8) comprising;

(1) receiving one or more previously determined and identified histograms generated from one or more previously determined first 2D image slices, spanning a three dimensional (3D) volume, where one or more components are identified in the one or more previously determined first 2D image slices where the one or more components include;

one or more of bone, blood vessel structures, vasculature, and blood vessels, where one or more connected components are used in identifying a plurality of pixels in the one or more components, where the plurality of pixels belong to the same one or more components received from the one or more previously determined first 2D image slices;

(2) receiving a user selected image slice from a plurality of second 2D image slices and then computing a histogram for the user selected 2D image slice;

(3) computing a cross correlation between one or more of the plurality of previously determined and identified histograms of step (1) and the histogram of step (2);

(4) identifying an anatomical structure in the user selected 2D image slice based on the cross correlation in step (3);

(5) assigning a region of the user selected 2D image slice corresponding to the anatomical structure identified in step (4) to one or more volumetric regions of the 3D volume;

(6) performing a threshold segmentation on the user selected 2D image slice;

(7) identifying one or more connected components of the anatomical structure identified in step (4) based on the results of the threshold segmentation of step (6); and

(8) displaying the threshold segmented 2D image slice, where the displayed threshold segmented 2D image slice is capable of producing one of each type of the following 2D image slices;

a bone slice, a blood vessel structure slice, a vasculature slice and a slice depicting blood vessels removed from the displayed threshold segmented 2D image slice based on the one or more connected components identified in step (7).

10. The system of claim 9, where the one or more connected components include both bone structures and vessel structures.

11. The system of claim 10, where the bone structures are removed from the displayed threshold segmented 2D image slice provided in step (8) by the digital data processing system.

12. The system of claim 11, where only the vessel structures are shown in the displayed threshold segmented 2D image slice provided from step (8) by the digital data processing system.

13. The system of claim 11, where the bone structures are automatically removed from the displayed threshold segmented 2D image slice provided from step (8) by the digital data processing system.

14. The system of claim 11, where the plurality of second 2D image slices are received from one or more previously performed computed tomography (CT) 3D run-off study or studies.

15. The system of claim 14, where the system automatically extracts one or more bone structures shown in the CT 3D run-off studies without removing one or more vessel structures in the displayed threshold segmented 2D image slice.

16. The system of claim 15, where the system automatically extracts the one or more bone structures in the previously performed CT 3D run-off studies, where the extracted one or more bone structures have the same or similar grey values to the one or more vessel structures that are shown in the displayed threshold segmented 2D image slice.

17. The system of claim 9, where step (3) involves computing with the digital data processing system a correlation between a region in the histogram of the 2D image slice of step (2) and a corresponding region in one or more of the previously determined and identified histograms generated from the plurality of 2D image slices, that is then provided and displayed to a user of the digital data processing system.

18. The system of claim 17, further comprising with the digital data processing system checking the assigning and identifying of steps (4), (5) and (7) with the 3D volume to insure that the assignment(s) and identification(s) are consistent with the imaged anatomy of the 3D volume.

19. A non-transitory method of processing anatomical structures present in two dimensional 2D image slices using a digital data processor that permits the identification of bone structures, vessel structures, or both bone structures and vessel structures, where bone structures are viewable without vessel structures, where vessel structures are viewable without bone structures, and where the method does not require the anatomical mid sagittal plane to be calculated within any of steps (a)-(f) comprising:

(a) selecting a region of a 2D image slice from a plurality of first 2D image slices spanning a three dimensional (3D) volume, using a digital data processing system configured for image processing;

(b) computing with the digital data processing system a histogram of the selected region of the 2D image slice;

(c) receiving one or more previously determined and identified histograms generated from one or more previously determined second 2D image slices, where one or more components are identified in the one or more previously determined second 2D image slices, wherein the one or more components include;

one or more of bone, blood vessel structures, vasculature, and blood vessels, where one or more connected components are used in identifying a plurality of pixels in the one or more components, where the plurality of pixels belong to the same one or more components received from the one or more previously determined second 2D image slices;

(d) computing with the digital data processing system a cross correlation between the histogram of step (b) and one or more of the plurality of previously determined and identified histograms generated from the second plurality of 2D image slices of step (c);

(e) identifying with the digital data processing system, two or more anatomical structures in the region of the 2D image slice of steps (a) and (b), based on the cross correlation computed in step (d);

(f) using the digital data processing system in order to;

(1) assign the selected region of the 2D image slice from steps (a) and (b), that corresponds to an anatomical structure of the two or more anatomical structures identified in step (e) to one or more corresponding volumetric regions of the 3D volume;

(2) perform a threshold segmentation on the 2D image slice of step (f)(1);

(3) identify one or more connected components of the two or more anatomical structures identified in step (e) based on the results of the threshold segmentation of step (2); and

(4) display the threshold segmented 2D image slice, where the displayed threshold segmented 2D image slice is capable of producing one of each type of the following 2D image slices;

a bone slice, a blood vessel structure slice, a vasculature slice and a slice depicting blood vessels removed from the displayed threshold segmented 2D image slice based on the one or more connected components identified in step (3).