PROBABILISTIC SEGMENTATION IN COMPUTER-AIDED DETECTION

• US 20100166276A1
• Filed: 09/08/2009
• Published: 07/01/2010
• Est. Priority Date: 11/11/2008
• Status: Active Grant

• Current Assignee: Siemens Medical Solutions USA Incorporated (Siemens AG)
• Original Assignee: Siemens Medical Solutions USA Incorporated (Siemens AG)
• Inventors: Dube, Shishir, Lakare, Sarang, Shinagawa, Yoshihisa, Valadez, Gerardo Hermosillo
• Granted Patent: US 8,265,390 B2
• Time in Patent Office: Days
• US Class Current: 382/131
• CPC Class Codes: G06T 2207/30068 Mammography; Breast
• CPC Class Codes: G06T 5/40 using histogram techniques
• CPC Class Codes: G06T 5/70 Denoising; Smoothing
• CPC Class Codes: G06T 7/11 Region-based segmentation
• CPC Class Codes: G06T 7/187 involving region growing; involving region merging; involving connected component labelling
• CPC Class Codes: G06V 10/267 by performing operations on regions, e.g. growing, shrinking or watersheds

Abstract

Described herein is a method and system for facilitating segmentation of images. A difference image is received and processed to extract at least one histogram (402). A noise component is determined by fitting a symmetric Gaussian distribution to the extracted histogram such that the negative portion of the Gaussian distribution coincides with the negative portion of the histogram (403). The noise component is then subtracted from the histogram to generate a probability distribution function (404), which may be converted to a cumulative distribution function (406) and applied to the difference image to generate a probabilistic representation of contrast enhancement (408).

Claims

1. A computer-implemented method of segmentation, comprising:

receiving at least one difference image;

extracting at least one histogram of the difference image;

determining a noise component by fitting a symmetric Gaussian distribution to the histogram such that a negative portion of the Gaussian distribution coincides with a negative portion of the histogram;

subtracting the noise component from the histogram to generate a probability distribution function;

converting the probability distribution function to a cumulative distribution function; and

applying the cumulative distribution function to the difference image to generate a probabilistic representation of contrast enhancement (I_CE).

2. The method of claim 1 further comprising:

generating the difference image by subtracting at least one baseline image from at least one contrast-enhanced image.

3. The method of claim 2 further comprising:

acquiring, by an imaging device, the baseline image or the contrast-enhanced image by one of a magnetic resonance (MR) imaging, computed tomographic (CT), helical CT, x-ray, positron emission tomographic, fluoroscopic, ultrasound and single photon emission computed tomographic (SPECT) technique.

4. The method of claim 2 wherein the contrast-enhanced image comprises an image of an anatomical part.

5. The method of claim 2 wherein the contrast-enhanced image comprises a dynamic contrast-enhanced magnetic resonance (MR) image.

6. The method of claim 1 further comprising:

outputting a probabilistic segmentation mask based on the probabilistic representation of contrast enhancement, wherein each voxel of the probabilistic segmentation mask is indicative of a likelihood that the voxel belongs to a pre-determined class based on at least one feature.

7. The method of claim 6 wherein the pre-determined class comprises a tumor class.

8. The method of claim 6 further comprising:

providing a seed point in the probabilistic representation of contrast enhancement (I_CE); and

growing a segmented region from the seed point using intensity properties from the probabilistic representation of contrast enhancement (I_CE).

9. The method of claim 8 further comprising providing the seed point via a one-click user interface.

10. The method of claim 8 further comprising providing the seed point by detecting a centroid, a most enhancing point or a most suspicious point in a segmented region-of-interest.

11. The method of claim 8 wherein the step of growing the segmented region from the seed point comprises:

computing at least one connectivity of a point in the probabilistic representation of contrast enhancement (I_CE) to the seed point.

12. The method of claim 11 further comprises computing I_CE(x, y, z)*I_CE(x₀, y_o, z_o), wherein (x, y, z) is a location of a point and (x₀, y₀, z₀) is a location of the seed point.

13. The method of claim 8 wherein the step of growing the segmented region from the seed point comprises:

computing at least one distance of a point in the probabilistic representation of contrast enhancement (I_CE) to the seed point.

14. The method of claim 13 wherein the distance comprises a non-negative distance.

15. The method of claim 14 wherein the distance comprises a Manhattan distance.

16. The method of claim 13 further comprising assigning, to a voxel of the probabilistic segmentation mask, a likelihood of one, if the distance is less than a distance threshold.

17. The method of claim 13 further comprising assigning, to a voxel of the probabilistic segmentation mask, a likelihood determined by a decreasing exponential function of the distance, if the distance is more than a distance threshold.

18. A computer usable medium having a computer readable program code tangibly embodied therein, said computer readable program code adapted to be executed by a processor to implement a method of segmentation, comprising:

receiving at least one difference image;

extracting at least one histogram of the difference image;

determining a noise component by fitting a symmetric Gaussian distribution to the histogram such that a negative portion of the Gaussian distribution coincides with a negative portion of the histogram;

subtracting the noise component from the histogram to generate a probability distribution function;

converting the probability distribution function to a cumulative distribution function; and

applying the cumulative distribution function to the difference image to generate a probabilistic representation of contrast enhancement (I_CE).

19. The computer usable medium of claim 18 wherein the method of segmentation further comprises:

outputting a probabilistic segmentation mask based on the probabilistic representation of contrast enhancement, wherein each voxel of the probabilistic segmentation mask is indicative of a likelihood that the voxel belongs to a pre- determined class based on at least one feature.

20. An image segmentation system, comprising:

a memory device for storing computer readable program code; and

a processor in communication with the memory device, the processor being operative with the computer readable program code to;

receive at least one difference image;

extract at least one histogram of the difference image;

determine a noise component by fitting a symmetric Gaussian distribution to the histogram such that a negative portion of the Gaussian distribution coincides with a negative portion of the histogram;

subtract the noise component from the histogram to generate a probability distribution function;

convert the probability distribution function to a cumulative distribution function; and

apply the cumulative distribution function to the difference image to generate a probabilistic representation of contrast enhancement (I_CE).