System and method for automatic bone extraction from a medical image

1. A method for automatically extracting a bone structure from a medical image, comprising:

using a processor to carry out the steps of:

performing a thresholding on each voxel of the medical image in an intensity range of the bone structure to generate a first bit mask;

eroding the first bit mask to remove connections between blood vessels and the bone structure to generate a second bit mask, wherein the connections are within the intensity range;

performing a region growing within the second bit mask starting from a seed point by growing voxels into neighboring voxels within the intensity range to separate the bone structure from unconnected blood vessels and to generate a third bit mask;

dilating the third bit mask to recover a surface area of the third bit mask to generate a fourth bit mask; and

performing a region growing within the fourth bit mask starting from the seed point by growing into neighboring voxels within the intensity range to generate a fifth bit mask representing the bone structure.

2. The method of claim 1 , wherein the medical image is of an anatomical part.

3. The method of claim 1 , wherein the intensity range is one of determined by selecting a brightness level of the bone structure, derived from an area surrounding the seed point, and preset.

4. The method of claim 1 , wherein the seed point is obtained by a user selecting a point representing an object to be removed from the medical image.

5. The method of claim 4 , wherein the object to be removed from the medical image is the bone structure.

6. The method of claim 1 , further comprising:

using the processor to carry out the step of:

performing a volume rendering technique on the fifth bit mask to extract the bone structure from the medical image.

7. The method of claim 6 , wherein the volume rendering technique is one of a maximum intensity projection, opacity rendering, and shaded opacity rendering.

8. The method of claim 1 , wherein the medical image is acquired by one of a computed tomography (CT) and helical CT imaging technique.

9. A system for automatic bone extraction from a medical image, comprising:

a memory device for storing a program;

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

perform a thresholding on each voxel of the medical image in an intensity range of the bone to generate a first bit mask;

erode the first bit mask to remove connections between blood vessels and the bone structure to generate a second bit mask, wherein the connections are within the intensity range;

perform a region growing within the second bit mask starting from a seed point by growing voxels into neighboring voxels within the intensity range to separate the bone from unconnected blood vessels and to generate a third bit mask;

dilate the third bit mask to recover a surface area of the third bit mask to generate a fourth bit mask; and

perform a region growing within the fourth bit mask starting from the seed point by growing into neighboring voxels within the intensity range to generate a fifth bit mask representing the bone.

10. The system of claim 9 , wherein the medical image is of an anatomical part.

11. The system of claim 9 , wherein the intensity range is one of determined by selecting a brightness level of the bone, derived from an area surrounding the seed point, and preset.

12. The system of claim 9 , wherein the seed point is obtained by a user selecting a point representing an object to be removed from the medical image.

13. The system of claim 12 , wherein the object to be removed from the medical image is the bone.

14. The system of claim 9 , wherein the processor is further operative with the program code to:

perform a volume rendering technique on the fifth bit mask to extract the bone from the medical image.

15. The system of claim 14 , wherein the volume rendering technique is one of a maximum intensity projection, opacity rendering, and shaded opacity rendering.

16. The system of claim 9 , wherein the medical image is acquired by one of a computed tomography (CT) and helical CT imaging technique.

17. A computer readable storage medium storing instructions that, when executed by a computer, cause the computer to perform a method of extracting a bone structure from a medical image, the method comprising:

performing a thresholding on each voxel of the medical image in an intensity range of the bone structure to generate a first bit mask;

eroding the first bit mask to remove connections between blood vessels and the bone structure to generate a second bit mask, wherein the connections are within the intensity range;

performing a region growing within the second bit mask starting from a seed point by growing voxels into neighboring voxels within the intensity range to separate the bone structure from unconnected blood vessels and to generate a third bit mask;

dilating the third bit mask to recover a surface area of the third bit mask to generate a fourth bit mask; and

performing a region growing within the fourth bit mask starting from the seed point by growing into neighboring voxels within the intensity range to generate a fifth bit mask representing the bone structure.

18. A system for automatic bone extraction from a medical image, comprising:

means for performing a thresholding on each voxel of the medical image in an intensity range of the bone to generate a first bit mask;

means for eroding the first bit mask to remove connections between blood vessels and the bone structure to generate a second bit mask, wherein the connections are within the intensity range;

means for performing a region growing within the second bit mask starting from a seed point by growing voxels into neighboring voxels within the intensity range to separate the bone from unconnected blood vessels and to generate a third bit mask;

means for dilating the third bit mask to recover a surface area of the third bit mask to generate a fourth bit mask; and

means for performing a region growing within the fourth bit mask starting from the seed point by growing into neighboring voxels within the intensity range to generate a fifth bit mask representing the bone.

19. A method for automatically extracting a bone from a slice-based three-dimensional (3D) medical image, comprising:

using a processor to carry out the steps of:

obtaining a bone seed point from the medical image;

determining an intensity range of the bone;

performing a binary thresholding on each voxel of the medical image in the intensity range to generate a first bit mask;

applying a morphological erosion filter to the first bit mask to remove connections between blood vessels and the bone to generate a second bit mask, wherein the connections are within the intensity range;

performing a region growing within the second bit mask starting from the seed point by growing voxels within the intensity range to separate the bone from unconnected blood vessels and to generate a third bit mask;

applying a dilation filter to the third bit mask to recover a surface area of the third bit mask to generate a fourth bit mask;

performing a region growing within the fourth bit mask starting from the seed point within the intensity range to generate a fifth bit mask representing the bone; and

performing a volume rendering technique on the fifth bit mask to extract the bone from the medical image.