System and method for automatic detection and localization of 3D bumps in medical images
View Patent ↗A method of detecting irregularities in a digitized image includes selecting a volume of interest in an image, growing a surface from a point in said volume of interest, determining the local gradient of each point on said surface, and applying a local filter to said gradient, determining a histogram of filtered gradients, and detecting an irregularity by selecting those points corresponding to gradient histogram values above a threshold value.
1. A computer implemented method of detecting irregularities in a digitized image, the method performed by said computer comprising the steps of:
providing a digitized image comprising a plurality of intensities corresponding to a domain of points on a 3-dimensional grid;
selecting a sub-volume of interest in said image;
growing a surface from a point in said volume of interest;
determining the local gradient of each point on said surface, and applying a local filter to said gradient;
determining a histogram of filtered gradients, and
detecting an irregularity by selecting those points corresponding to gradient histogram values above a threshold value.
2. The method of claim 1 , wherein said volume of interest is selected by a user.
3. The method of claim 1 , wherein said image is subdivided into a plurality of sub-volumes, wherein each sub-volume is a volume of interest.
4. The method of claim 1 wherein said surface is grown based on an intensity constraint and a spatial constraint.
5. The method of claim 4 , wherein said intensity constraint comprises a threshold wherein points on said surface are selected, and said spatial constraint comprises at least one neighbor of said surface point that is also a surface point.
6. The method of claim 1 , wherein said gradient is computed over a 3×3×3 neighborhood of each surface point, and wherein said filter comprises a 3×3×3 operator defined for each of three axes, and further comprising summing each of said three filters for each surface point.
7. The method of claim 6 , wherein said histogram relates each surface point to its filter sum for each of said three axes.
8. The method of claim 1 , wherein said histogram threshold value is determined by applying a Laplacian operator to analyze differences in said histogram values.
9. The method of claim 6 wherein said filter comprises an X-axis operator on a plane perpendicular to a Z-axis, an Y-axis operator on a plane perpendicular to said Z-axis, and a Z-axis operator on a plane perpendicular to said Z-axis, wherein said X-axis operator takes the form
−1
0
1
−1
0
1
−1
0
1
−1
0
1
−1
0
1
−1
0
1
−1
0
1
−1
0
1
−1
0
1
said Y-axis operator takes the form
1
1
1
1
1
1
1
1
1
0
0
0
0
0
0
0
0
0
−1
−1
−1
−1
−1
−1
−1
−1
−1
and said Z-axis operator takes the form
−1
−1
−1
0
0
0
1
1
1
−1
−1
−1
0
0
0
1
1
1
−1
−1
−1
0
0
0
1
1
1.
10. A computer implemented method of detecting irregularities in a digitized image, the method performed by said computer comprising the steps of:
determining a gradient of each point on a surface of interest in said image, and applying a local filter to said gradient, wherein said gradient is computed over a 3×3×3 neighborhood of each surface point, and wherein said filter comprises a 3×3×3 operator defined for each of three axes, and further comprising summing each of said three filters for each surface point;
determining a histogram of filtered gradients that relates each surface point to its filter sum for each of said three axes; and
detecting an irregularity by selecting those points corresponding to gradient histogram values above a threshold value.
11. The method of claim 10 , further comprising selecting a volume of interest in said image, and growing said surface from a point in said volume of interest.
12. A program storage device readable by a computer, tangibly embodying a program of instructions executable by the computer to perform the method steps for detecting irregularities in a digitized image, said method comprising the steps of:
providing a digitized image comprising a plurality of intensities corresponding to a domain of points on a 3-dimensional grid;
selecting a sub-volume of interest in said image;
growing a surface from a point in said volume of interest;
determining the local gradient of each point on said surface, and applying a local filter to said gradient;
determining a histogram of filtered gradients, and
detecting an irregularity by selecting those points corresponding to gradient histogram values above a threshold value.
13. The computer readable program storage device of claim 12 , wherein said volume of interest is selected by a user.
14. The computer readable program storage device of claim 12 , wherein said image is subdivided into a plurality of sub-volumes, wherein each sub-volume is a volume of interest.
15. The computer readable program storage device of claim 12 , wherein said surface is grown based on an intensity constraint and a spatial constraint.
16. The computer readable program storage device of claim 15 , wherein said intensity constraint comprises a threshold wherein points on said surface are selected, and said spatial constraint comprises at least one neighbor of said surface point that is also a surface point.
17. The computer readable program storage device of claim 12 , wherein said gradient is computed over a 3×3×3 neighborhood of each surface point, and wherein said filter comprises a 3×3×3 operator defined for each of three axes, and further comprising summing each of said three filters for each surface point.
18. The computer readable program storage device of claim 17 , wherein said histogram relates each surface point to its filter sum for each of said three axes.
19. The computer readable program storage device of claim 12 , wherein said histogram threshold value is determined by applying a Laplacian operator to analyze differences in said histogram values.
20. The computer readable program storage device of claim 17 , wherein said filter comprises an X-axis operator on a plane perpendicular to a Z-axis, an Y-axis operator on a plane perpendicular to said Z-axis, and a Z-axis operator on a plane perpendicular to said Z-axis, wherein said X-axis operator takes the form
−1
0
1
−1
0
1
−1
0
1
−1
0
1
−1
0
1
−1
0
1
−1
0
1
−1
0
1
−1
0
1
said Y-axis operator takes the form
1
1
1
1
1
1
1
1
1
0
0
0
0
0
0
0
0
0
−1
−1
−1
−1
−1
−1
−1
−1
−1
and said Z-axis operator takes the form
−1
−1
−1
0
0
0
1
1
1
−1
−1
−1
0
0
0
1
1
1
−1
−1
−1
0
0
0
1
1
1.