Method for segmenting objects in images

Patent metadata

Field	Value
Publication number	US-9830698-B2
Application number	US-201414480505-A
Country	US
Kind code	B2
Filing date	Sep 8, 2014
Priority date	May 21, 2007
Publication date	Nov 28, 2017
Grant date	Nov 28, 2017

Abstract

A method for identifying an attribute of an object represented in an image comprising data defining a predetermined spatial granulation for resolving the object, where the object is in contact with another object. In an embodiment, the method comprises identifying data whose values indicate they correspond to locations completely within the object, determining a contribution to the attribute provided by the data, and identifying additional data whose values indicate they are not completely within the object. The method next interpolates second contributions to the attribute from the values of the additional data and finds the attribute of the object from the first contribution and second contributions. The attribute may be, for example, a volume, and the values may correspond, for example, to intensity.

First claim

We claim:

1. A computer-implemented image processing method, the method comprising:
receiving, by an image processor, a magnetic resonance imaging (MRI) image from an MRI system, wherein the MRI image comprises a plurality of voxels;
identifying, by the image processor, one or more partial voxels corresponding to a region of the MRI image that is not completely filled with blood, wherein the one or more partial voxels are located at an edge of a first object and wherein the one or more partial voxels encompass a portion of the first object and a portion of a second object, wherein the first object and the second object are adjacent to one another; and
generating, by the image processor, an estimated value corresponding to an estimate of an amount or proportion of the first object corresponding to the one or more partial voxels based on an attribute of the one or more partial voxels; and
outputting, by the image processor, the estimated value to a medical provider, so as to facilitate diagnosis and/or treatment of a medical condition.

2. The computer-implemented method according to claim 1 , further comprising:
correcting the MRI image to account for sensitivity conditions.

3. The computer-implemented method according to claim 1 , wherein partial volume interpolation is used to generate the estimated value.

4. The computer-implemented method according to claim 1 , wherein intensities of the one or more partial voxels correspond to the attribute of the one or more partial voxels.

5. A non-transitory processor-readable medium having processor-executable instructions stored thereon for image processing, the processor-executable instructions, when executed by an image processor, facilitating the performance of the following:
receiving, by an image processor, a magnetic resonance imaging (MRI) image from an MRI system, wherein the MRI image comprises a plurality of voxels;
identifying, by the image processor, one or more partial voxels corresponding to a region of the MRI image that is not completely filled with blood, wherein the one or more partial voxels are located at an edge of a first object and wherein the one or more partial voxels encompass a portion of the first object and a portion of a second object, wherein the first object and the second object are adjacent to one another;
generating, by the image processor, an estimated value corresponding to an estimate of an amount or proportion of the first object corresponding to the one or more partial voxels based on an attribute of the one or more partial voxels; and
outputting, by the image processor, the estimated value to a medical provider, so as to facilitate diagnosis and/or treatment of a medical condition.

6. The non-transitory processor-readable medium according to claim 5 , wherein the processor-executable instructions further facilitate the performance of the following:
correcting the MRI image to account for sensitivity conditions.

7. The non-transitory processor-readable medium according to claim 5 , wherein partial volume interpolation is utilized to generate the estimated value.

8. The non-transitory processor-readable medium according to claim 5 , wherein intensities of the one or more partial voxels correspond to the attribute of the one or more partial voxels.

Assignees

• Univ Cornell
• Univ Cornell

Inventors

• Wang Yi
• Codella Noel C F
• Lee Hae-Yeou

Classifications

• G06T7/11 Primary

  Region-based segmentation · CPC title

• G06T2207/10072

  Tomographic images · CPC title

• G06T7/174

  involving the use of two or more images · CPC title

• G06T7/62

  of area, perimeter, diameter or volume · CPC title

• G06T7/0012 Primary

  Biomedical image inspection · CPC title