System and method for detecting a protrusion in a medical image

1. A method for detecting a protrusion in a medical image, comprising:

segmenting a medical image;

calculating a distance map of the medical image;

calculating a gradient of the distance mapped medical image; and

processing the gradient to detect a protrusion in the medical image, wherein the gradient is processed by:

projecting a first plurality of rays from a location in the distance mapped medical image;

calculating a value for each of the first plurality of rays based on features of each of the first plurality of rays and the gradient of the distance mapped medical image;

summing and scaling the value of each of the first plurality of rays; and

detecting one of a sphere-like and polyp-like shape using the summed and scaled values of the first plurality of rays, wherein one of the sphere-like and polyp-like shapes is the protrusion.

2. The method of claim 1 , farther comprising:

acquiring the medical image.

3. The method of claim 2 , wherein the medical image is acquired by one of a computed tomographic (CT), helical CT, x-ray, positron, emission tomogaphic, fluoroscopic, ultrasound, and magnetic resonance (MR) imaging technique.

4. The method of claim 2 , wherein the medical image is of an anatomical part.

5. The method of claim 1 , wherein the processing step further comprises:

projecting a second plurality of rays from a location comprising an original distance value in the distance mapped medical image;

calculating an absolute value of a difference between a length of each of the second plurality of rays and a distance value at an end of each of the second plurality of rays, wherein the length of each of the second plurality of rays is a fraction of the original distance value from the location;

dividing a sum of the absolute value by the total number of the second plurality of rays; and

detecting one of a sphere-like and polyp-like shape using the division result, wherein one of the sphere-like and polyp-like shapes is the protrusion.

6. The method of claim 1 , wherein the processing step further comprises:

projecting a second plurality of rays from a location comprising an original distance value in the distance mapped medical image;

determining a distance value for each of the second plurality of rays that is a fraction of the original distance value from the location;

calculating a sphere-based response, wherein the sphere-based response is calculated by:

∑

i

∈

S

⁢

(

d

-

l

i

)

T

where d is the original distance value, l i is the length of a ray i, T is a total number of the second plurality of rays, and S is a set of the second plurality of rays such that l i <d; and

detecting the protrusion using the sphere-based response.

7. The method of claim 1 , wherein the processing step further comprises:

projecting a second plurality of rays from a location comprising an original distance value in the distance mapped medical image;

determining a distance value for each of the second plurality of rays that has a supplementary ray that has a distance value less than the original distance value;

calculating a hemisphere-based response; wherein the hemisphere-based response is calculated by:

∑

i

∈

S

⁢

(

d

-

l

i

)

T

/

2

where d is the original distance value, l i is the length of a ray i, T is a total number of the second plurality of rays, and S is a set of the second plurality of rays whose supplementary rays do hot have a value less than the original distance value; and

detecting the protrusion using the hemisphere-based response.

8. The method of claim 1 , wherein the processing step further comprises:

projecting a second plurality of rays from an edge of the distance mapped medical image, wherein the second plurality of rays follow the steepest gradient; and

accumulating paths of the second plurality of rays, wherein the accumulated paths form a response image for detecting the protrusion.

9. The method of claim 1 , wherein the processing step further comprises:

projecting a second plurality of rays from a location comprising an original distance value in the distance mapped medical image;

determining a distance value for each of the second plurality of rays that is a fraction of the original distance value from the location;

calculating a sphere-based response, wherein the sphere-based response is calculated by:

∑

i

=

0

F

×

d

⁢



d

i

-

l

i



T

where d is the original distance value, F is a fractional value between 0 and 1, dis the distance value at a point along one of the second plurality of rays, l i is the length of one of the second plurality of rays at a point i, and T is the total number of points taken from i=0 to i=F*d;

calculating a gray-level difference of the distance mapped medical image, wherein the gray level difference is calculated by:

∑

i

=

0

K

⁢

r

k

K

where rk represents the sphere-based response for each ray k; and

detecting the protrusion using the gray-level difference.

10. The method of claim 1 , wherein the protrusion is one of a nodule, lesion, polyp, pre-cancerous growth, and cancerous growth.

11. The method of claim 1 , further comprising:

storing a list of one or more detected protrusions; and

filtering one or more false positives from the list, wherein one of the false positives is not one of a nodule, lesion, polyp, pre-cancerous growth, and cancerous growth.

12. A system for detecting a protrusion in a medical image, comprising:

a memory device for storing program code;

a processor in communication with the memory device, the processor operative with the program code to:

segment a medical image;

calculate a distance map of the medical image;

calculate a gradient of the distance mapped medical image; and

process the gradient to detect a protrusion in the medical image, wherein when processing the gradient, the processor is operative with the program code to:

project a first plurality of rays from a location in the distance mapped medical image;

calculate a value for each of the first plurality of rays based on features of each of the first plurality of rays and the gradient of the distance mapped medical image;

summing and scaling the value for each of the first plurality of rays; and

detecting one of a sphere-like and polyp-like shape using the summed and scaled values of the first plurality of rays, wherein one of the sphere-like and polyp-like shapes is the protrusion.

13. The system of claim 12 , wherein the processor is further operative with the program code to:

acquire the medical image, wherein the medical image is of an anatomical part.

14. The system of claim 13 , wherein the medical image is acquired by one of a computed tomographic (CT), helical CT, x-ray, positron emission tomographic, fluoroscopic, ultrasound, and magnetic resonance (MR) imaging technique.

15. The system of claim 12 , wherein the processor is further operative with the program code when processing the gradient to:

project a second plurality of rays from a location comprising an original distance value in the distance mapped medical image;

calculate an absolute value of a difference between a length of each of the second plurality of rays and a distance value at an end of each of the second plurality of rays, wherein the length of each of the second plurality of rays is a fraction of the original distance value from the location;

divide a sum of the absolute value by the total number of the second plurality of rays; and

detect one of a sphere-like and polyp-like shape using the division result, wherein one of the sphere-like and polyp-like shapes is the protrusion.

16. The system of claim 12 , wherein the processor is further operative with the program code when processing the gradient to:

project a second plurality of rays from a location comprising an original distance value in the distance mapped medical image;

determine a distance value for each of the second plurality of rays that is a fraction of the original distance value from the location;

calculate a sphere-based response of the second plurality of rays;

calculate a hemisphere-based response of the second plurality of rays; and

detect the protrusion using the sphere and hemisphere-based responses.

17. The system of claim 12 , wherein the processor is further operative with the program code when processing the gradient to:

project a second plurality of rays from an edge of the distance mapped medical image, wherein the second plurality of rays follow the steepest gradient; and

accumulate paths of the second plurality of rays, wherein the accumulated paths form a response image for detecting the protrusion.

18. The system of claim 12 , wherein the processor is further operative with the program code when processing the gradient to:

project a second plurality of rays from a location comprising an original distance value in the distance mapped medical image;

determine a distance value for each of the second plurality of rays that is a fraction of the original distance value from the location;

calculate a sphere-based response of the second plurality of rays;

calculate a gray-level difference of the distance mapped medical image; and

detect to protrusion using to sphere-based response and the gray-level difference.

19. The system of claim 12 , wherein the protrusion is one of a nodule, lesion, polyp, pre-cancerous growth, and cancerous growth.

20. The system of claim 12 , wherein the processor is further operative with the program code when processing the gradient to:

store a list of one or more detected protrusions; and

filter one or more false positives from the list, wherein one or more of the false positives is not one of a nodule, lesion, polyp, pre-cancerous growth, and cancerous growth.

21. A computer program product comprising a computer useable medium having computer program logic recorded thereon for detecting a protrusion in a medical image, the computer program logic comprising:

program code for segmenting a medical image;

program code for calculating a distance map of the medical image;

program code for calculating a gradient of the distance mapped medical image; and

program code for processing the gradient to detect a protrusion in the medical image, wherein the gradient is processed by;

projecting a plurality of rays from a location in the distance mapped medical image;

calculating a value for each of the plurality of rays based on features of each of the plurality of rays and the gradient of the distance mapped medical image;

summing and scaling the value of each of the plurality of rays; and

detecting one of a sphere-like and polyp-like shape using the summed and scaled values of the plurality of rays, wherein one of the sphere-like and polyp-like shapes is the protrusion.

22. The computer program product of claim 21 , further comprising:

program code for acquiring the medical image.

23. The system computer program product of claim 22 , wherein the image is acquired by one of a computed tomogaphic (CT), helical CT, x-ray, positron emission tomographic, fluoroscopic, ultrasound, and magnetic resonance (MR) imaging technique.

24. The computer program product of claim 21 , wherein the protrusion is one of a nodule, lesion, polyp, pre-cancerous growth, and cancerous growth.

25. A method for detecting a protrusion in a medical image, comprising:

segmenting a medical image;

calculating a distance map of the medical image;

calculating a gradient of the distance mapped medical image; and

processing the gradient to detect a protrusion in the medical image, wherein the gradient is processed by:

projecting a plurality of rays from a location comprising an original distance value in the distance mapped medical image;

calculating an absolute value of a difference between a length of each of the plurality of rays and a distance value at an end of each of the plurality of rays, wherein the length of each of the plurality of rays is a fraction of the original distance value from the location;

dividing a sum of the absolute value by the total number of the plurality of rays; and

detecting one of a sphere-like and polyp-like shape using the division result, wherein one of the sphere-like and polyp-like shapes is the protrusion.

26. A method for detecting a protrusion in a medical image, comprising:

segmenting a medical image;

calculating a distance map of the medical image;

calculating a gradient of the distance mapped medical image; and

processing the gradient to detect a protrusion in the medical image, wherein the gradient is processed by using first, second or third techniques:

wherein the first technique includes:

projecting a plurality of rays from a location comprising an original distance value in the distance mapped medical image;

determining a distance value for each of the plurality of rays that is a fraction of the original distance value from the location;

calculating a sphere-based response of the plurality of rays; and

detecting the protrusion using the sphere-based response;

wherein the second technique includes:

projecting plurality of rays from a location comprising an original distance value in the distance mapped medical image;

determining a distance value for each of the plurality of rays that is a fraction of the original distance value from the location;

calculating a sphere-based response of the plurality of rays;

calculating a gray-level difference of the distance mapped medical image; and

detecting the protrusion using the gray-level difference;

wherein the third technique includes:

projecting a plurality of rays from a location comprising an original distance value in the distance mapped medical image;

determining a distance value for each of the plurality of rays that has a supplementary ray that has a distance value less than the original distance value;

calculating a hemisphere-based response of the plurality of rays; and

detecting the protrusion using the hemisphere-based response.