Nodule detection

1. A method of identifying a nodule in a computed tomography scan image of at least part of a lung, comprising:

a. calculating a sphericity index map of a plurality of points within the image;

b. identifying a region of high sphericity within the image by identifying connected ones of said points having a sphericity index above a predetermined high sphericity index threshold as belonging to said region;

c. performing three-dimensional region growing from said region by adding to said region ones of said points having a sphericity index above a relaxed sphericity index threshold, said relaxed sphericity index threshold being lower than said high sphericity index threshold, to form an extended region; and

d. outputting, said extended region as an identified nodule.

2. The method of claim 1 , wherein the sphericity index indicates the variation in curvature of an iso-intensity surface in the region around each said point.

3. The method of claim 2 , wherein the sphericity index is calculated from a partial derivative of intensity around each said point.

4. The method of claim 3 , wherein the sphericity index is calculated from the first and second partial derivatives of intensity of said image in three dimensions.

5. The method of claim 1 , wherein the sphericity index map is calculated on a smoothed said image.

6. The method of claim 1 , further including performing filtering around said extended region using a spherical filter.

7. A method of identifying a nodule in a computed tomography scan image of at least part of a lung, comprising:

a. identifying a region of high sphericity within the image;

b. extending the region to include connected points of lower sphericity;

c. performing filtering around the extended region using a spherical filter comprising an inner spherical region of positive weight and an outer spherical region of negative weight, to form a filtered extended region; and

d. outputting said filtered extended region as an identified nodule.

8. The method of claim 7 , wherein the inner and outer spherical regions have substantially equal volumes.

9. The method of claim 7 , wherein the inner spherical region has a diameter substantially equal to the diameter of the extended region.

10. The method of claim 7 , wherein the filtering step comprises convolving the inner and outer spherical regions with said image.

11. The method of claim 10 , further comprising varying a diameter of the inner spherical region so as to determine a maximum strength of said convolution, and applying spherical filtering corresponding to the maximum convolution strength to the scan image to generate a spherically enhanced output image.

12. A method of identifying a nodule in a computed tomography scan image of at least part of a lung, comprising:

a. calculating a sphericity index map of a plurality of points within the image;

b. identifying a region of high sphericity within the image by identifying connected ones of said points having a sphericity index above a predetermined high sphericity index threshold as belonging to said region;

c. extending the region to include connected points of lower sphericity, to form an extended region;

d.performing filtering around the extended region using a spherical filter comprising an inner spherical region of positive weight and an outer spherical region of negative weight, wherein the filtering comprises convolving the inner and outer spherical regions with the sphericity index map, to form a filtered extended region; and

e. outputting the filtered extended region as an identified nodule.

13. The method of claim 12 , further comprising varying a diameter of the inner spherical region so as to determine a maximum strength of said convolution, and applying spherical filtering corresponding to the maximum convolution strength to the sphericity index map to generate a spherically enhanced output image.

14. Apparatus for identifying a nodule in a computed tomography scan image of at least part of a lung, comprising:

means for identifying a region of high sphericity within the image, including means for calculating a sphericity index map of a plurality of points within the image, and for identifying connected points having a sphericity index above a predetermined high sphericity index threshold as belonging to said region;

means for extending the region by performing three-dimensional region growing from said region to add connected points having a sphericity index above a relaxed sphericity index threshold, said relaxed sphericity index threshold being lower than said high sphericity index threshold, to form an extended region; and

means for outputting the extended region as an identified nodule.

15. The apparatus of claim 14 , wherein the sphericity index indicates the variation in curvature of an iso-intensity surface in the region around each said point.

16. The apparatus of claim 15 , wherein the sphericity index is calculated from a partial derivative of intensity around each said point.

17. The apparatus of claim 16 , wherein the sphericity index is calculated from the first and second partial derivatives of intensity of said image in three dimensions.

18. The apparatus of claim 14 , wherein the sphericity index map is calculated on a smoothed said image.

19. Apparatus for identifying a nodule in a computed tomography scan image of at least part of a lung, comprising:

means for calculating a sphericity index map of a plurality of points within the image;

means for identifying a region of high sphericity within the image by identifying connected ones of said points having a sphericity index above a predetermined high sphericity index threshold as belonging to said region;

means for performing three-dimensional region growing from said region to add connected points of lower sphericity, to form an extended region;

means for performing filtering around the extended region using a spherical filter comprising an inner spherical region of positive weight and an outer spherical region of negative weight, wherein the filtering comprises convolving the inner and outer spherical regions with the sphericity index map, to form a filtered extended region; and

means for outputting the filtered extended region as an identified nodule.

20. A computer readable storage medium having recorded thereon program code to perform the following processes for identifying a nodule in a computed tomography scan image of at least part of a lung:

a. calculating a sphericity index map of a plurality of points within the image;

b. identifying a connected region of high sphericity index within the image by identifying connected ones of said points having a sphericity index above a predetermined high sphericity index threshold as belonging to said connected region;

c. performing three-dimensional region growing from said connected region to add connected points of lower sphericity, to form an extended region;

d. performing filtering around the extended region using a spherical filter comprising an inner spherical region of positive weight and an outer spherical region of negative weight, wherein the filtering comprises convolving the inner and outer spherical regions with the sphericity index map, to form a filtered extended region; and

e. outputting the filtered extended region as an identified nodule.

21. An article comprising a computer readable storage medium for storing instructions to enable a processor-based system to:

a. calculate a sphericity index map of a plurality of points within a computed tomography scan image of at least part of a lung;

b. identify a region of high sphericity within the image by identifying connected ones of said points having a sphericity index above a predetermined high sphericity index threshold as belonging to said region;

c. extend the region to include connected points of lower sphericity, to form an extended region;

d. perform filtering around the extended region using a spherical filter comprising an inner spherical region of positive weight and an outer spherical region of negative weight, wherein the filtering comprises convolving the inner and outer spherical regions with the sphericity index map, to form a filtered extended region; and

e. output the filtered extended region as an identified nodule.

22. A computer readable storage medium having recorded thereon program code to perform the following processes for identifying a nodule in a computed tomography scan image of at least part of a lung:

a. calculating a sphericity index map of a plurality of points within the image;

b. identifying a region of high sphericity within the image by identifying connected ones of said points having a sphericity index above a predetermined high sphericity index threshold as belonging to said region;

c. performing three-dimensional region growing from said region by adding to said region ones of said points having a sphericity index above a relaxed sphericity index threshold, said relaxed sphericity index threshold being lower than said high sphericity index threshold, to form an extended region; and

d. outputting said extended region as an identified nodule.

23. An article comprising a computer readable storage medium for storing instructions to enable a processor-based system to:

a. calculate a sphericity index map of a plurality of points within a computed tomography scan image of at least part of a lung;

b. identify a region of high sphericity within the image by identifying connected ones of said points having a sphericity index above a predetermined high sphericity index threshold as belonging to said region;

c. perform three-dimensional region growing from said region by adding to said region ones of said points having a sphericity index above a relaxed sphericity index threshold, said relaxed sphericity index threshold being lower than said high sphericity index threshold, to form an extended region; and

d. output said extended region as an identified nodule.