Segmenting a series of 2D or 3D images

• US 7,668,370 B2
• Filed: 09/05/2003
• Issued: 02/23/2010
• Est. Priority Date: 09/19/2002
• Status: Active Grant

• Current Assignee: Koninklijke Philips Electronics N.V. (Koninklijke Philips N.V.)
• Original Assignee: Koninklijke Philips Electronics N.V. (Koninklijke Philips N.V.)
• Inventors: Noble, Nicholas Michael Ian, Spreeuwers, Lieuwe Jan, Breeuwer, Marcel
• Primary Examiner(s): Mehta; Bhavesh M
• Assistant Examiner(s): ZARKA, DAVID PETER
• Application Number:
  US10/527,862
• Publication Number:
  US 20050271271A1
• Time in Patent Office: 2,363 Days
• Field of Search: 382/173
• US Class Current: 382/173
• CPC Class Codes: G06T 2207/10072 Tomographic images
• CPC Class Codes: G06T 2207/30061 Lung
• CPC Class Codes: G06T 7/11 Region-based segmentation

Abstract

The invention relates to an apparatus for segmenting a series of 2D or 3D images obtained by monitoring a patient'"'"'s organ or other body part, wherein a first segmentation is carried out on a first image of the series of images and wherein the first segmentation is used for the subsequent segmentation of the remainder of images of the series of images. A series of transformations are carried out wherein each separate transformation embodies a fitting operation between two images of the series of images, and wherein substantially all images of the series of images are subject of such a transformation. The first segmentation on the first image of the series of images is modified and subsequently applied to any further image of the series of images according to the transformation or sequence of transformations that fits the first image to a further image of the series of images.

Claims

1. An apparatus for segmenting a first series of 2D or 3D images obtained from a target object within a patient, comprising:

a transform calculator which calculates a series of transformations, wherein each transformation comprises an operation for defining a best fit between two images of the first series of images, the transformations including;

(1) a first transformation of the series of transformations between a first image of the series and a subsequent second image of the series and (2) a second transformation between the first and a subsequent third image;

a segmenter which segments the first image of the first series is generated;

an image converter which applies the first transformation to the segmentation of the first image to generate a second image segmentation corresponding to the second image and applies the second transformation to the first segmentation to generate a third image segmentation corresponding to the third images,

wherein the segmentation of the first series of images is applied to a second series of images.

2. The apparatus according to claim 1, wherein each transformation relates one of the series of images to an adjacent one of the images of the first series of images.

3. The apparatus according to claim 1, wherein each of the first and second series of images are each collected form one or more of:

a magnetic resonance (MR) imaging system, a computer tomography (CT) imaging system, a nuclear medicine (NM) imaging system and an ultrasound (US) imaging system.

4. The apparatus according to claim 1, wherein the first and second series of images are collected at different times.

5. The apparatus according to claim 1, wherein the images relate to a sphere-like organ and prior to establishing the first series of transformations, the first series of images is converted to a modified first series of images showings walls of the organ in a flat plane wherein opposing sides of said plane correspond to an inside and an outside of said organ, and that the said series of transformations are applied to the modified first series of images.

6. A method for segmenting a first series of 2D or 3D images obtained from a target object within a patient, the method comprising:

with a processor, calculating a series of transformations in relation to the first series of images including at least;

(1) a first transformation of the series of transformations best fitting a first of the images and a subsequent second one of the images and (2) a second transformation of the series of transformation best fitting the first image and a subsequent third one of the images;

performing a segmentation on the first image of the first series of images to obtain a first segmented image;

with a processor, applying the first transformation to the segmentation of the first image to generate a second segmentation corresponding to second of the images and applying the second transformation to the segmentation of the first image to generate a third segmentation corresponding to the third image; and

at least one of storing the second and third segmented images in a memory and displaying the second and third segmented images on a display device,

wherein the segmentation of the first series of images is applied to a second series of images.

7. The method according to claim 6, further comprising:

collecting each of the first and second series of images with a different one of;

magnetic resonance (MR) imaging system, a computer tomography (CT) imaging, a nuclear medicine (NM) imaging and ultrasound (US) imaging.

8. The method according to claim 6, wherein the first and second series of images are collected at different times.

9. The method according to claim 6, wherein the images relate to a sphere-like organ, the method further comprising:

prior to establishing the series of transformations, converting the first series of images to a modified first series of images showings walls of the organ in a flat plane wherein opposing sides of said plane substantially correspond to an inside and an outside of said organ, and wherein the series of transformations are applied to the modified first series of images.

10. A method for segmenting a first series of 2D or 3D images, the method comprising:

with at least one processor, calculating a transformation between a first image and a second image of the first series of images to determine a first transformation of a series of transformations that best fits the first image and the second image;

performing a first segmentation of the first image of the first series of images to obtain a first segmented image according to a selected segmentation process;

with at least one processor, applying the first transformation to the first segmented image to generate a second segmented image corresponding to the second image;

with the at least one processor, calculating a second transformation between the first image and a third image of the first series of images which second transformation best fits the first image and the third image with the processor;

with the at least one processor, applying the second transformation to the first segmented image and to generate a third segmented image corresponding to the third image; and

at least one of storing the second and third segmented images in a memory and displaying the second and third segmented images on a monitors,

wherein the segmentation of the first series of images is applied to a second series of images.

11. The method of claim 10, further comprising:

calculating a series of the transformations among the first series of images prior to segmenting the first image,

wherein each image of the series of images comprises a wall of an organ in a flat plane, opposing sides of the wall respectively corresponding to inside and outside the organ.

12. The method of claim 11, further including resampling the first series of images.

13. A method for segmenting a first series of 2D or 3D images, the method comprising:

with one or more processors, calculating a first transformation between a first image and a second image of the first series of images to determine a first transformation of a series of transformations that best fits the first image and the second image;

with the one or more processors, calculating a second transformation between the first image and a successive third image of the first series of images which second transformation best fits the first image and the third image;

with the one or more processors, performing a first segmentation of the first image of the first series of images to obtain a first segmented image according to a selected segmentation process;

with the one or more processors, applying the first transformation to the first segmented image to generate a second segmented image corresponding to the second image;

with the one or more processors, applying the second transformation to the first segmented image to generate a third segmented image corresponding to the third image;

at least one of storing the segmented images in a memory and displaying the second segmented images on a display device,

wherein the segmentation of the first series of images is applied to a second series of images.