System and method for automatic detection of internal structures in medical images

1. A method for using a medical imaging system to recognize an internal structure from a three-dimensional image comprising a plurality of image sub-volumes, the method comprising:

selecting an image sub-volume of the plurality of image sub-volumes using a binary search pattern comprising:

identifying a middle image sub-volume of the plurality of image sub-volumes and using the identified middle image sub volume as a first-selected image sub-volume;

dividing the plurality of sub-volumes into a top section and a bottom section that are separated by the middle image sub-volume;

identifying a middle image sub-volume of the top section and using the identified middle image sub volume of the top section as a second-selected image sub-volume;

identifying a middle image sub-volume of the bottom section and using the identified middle image sub volume of the bottom section as a third-selected image sub-volume; and

recursively performing the above steps of dividing and identifying to find subsequent selected image sub-volumes;

analyzing the selected image sub-volume for the presence of the internal structure;

repeating the steps of selecting an image sub-volume using the binary search pattern and analyzing the selected sub-volume for the presence of the internal structure until the internal structure is found in an image sub-volume; and

identifying bounds of the internal structure based on the location of the image sub-volume within which the internal structure is found.

2. The method of claim 1 , wherein the medical imaging system is a computer tomography (CT) system or a magnetic resonance imaging (MRI) system.

3. The method of claim 1 , wherein the sub-volumes are image slices.

4. The method of claim 1 , wherein the internal structure is a colon, a lung or a heart.

5. The method of claim 1 , additionally comprising analyzing the three-dimensional image within the identified bounds of the internal structure to determine whether a contrast is present within the internal structure.

6. The method of claim 5 , wherein, when it is determined that contrast is present, it is determined whether the internal structure has an elevated risk of irregularity based on the contrast.

7. The method of claim 5 , wherein the step of determining whether a contrast is present within the internal structure comprises:

identifying a middle sub-volume of the internal structure based on the identified bounds;

identifying a region of interest within the internal structure;

detecting voxels with contrast within the region of interest by applying a predetermined threshold;

performing connected components labeling to define a region of contrast;

computing an area of defined region of contrast;

comparing an area of defined region of contrast against a minimum size criteria; and

determining contrast is present within the internal structure when the defined area of contrast is greater than the minimum size criteria.

8. The method of claim 1 , additionally comprising analyzing the three-dimensional image within the identified bounds of the internal structure to determine whether the internal structure has an elevated risk of irregularity.

9. The method of claim 8 , wherein the elevated risk of irregularity is a suspected emboli or cancer.

10. The method of claim 1 , wherein the step of analyzing the selected image sub-volume for the presence of the internal structure comprises:

performing a 3D to 2D transformation on the selected image sub-volume by extracting opacity values along rays on the image sub-volume plane, processing the rays by discarding opacity values greater than a fixed value, and computing the means of all opacity values along each ray;

generating a 1D profile on the selected image sub-volume by performing Gaussian smoothing and calculating column-wise average of the 2D transformation; and

analyzing the 1D profile against known 1D profiles of possible internal structures to determine the presence of the internal structure within the image sub-volume.

11. A method for determining whether a contrast is present within an internal structure identified from a three-dimensional image comprising a plurality of image sub-volumes, the method comprising:

identifying a middle sub-volume of the internal structure based on identified bounds of the internal structure, the identifying of the middle sub-volume including selecting an image sub-volume of the plurality of image sub volumes using a binary search pattern, determining whether the selected image sub-volume is the middle sub-volume of the internal structure, and repeating the selecting of the sub-volume using the binary search pattern and determining whether the selected image sub-volume is the middle sub-volume until the middle sub-volume is identified;

identifying a region of interest within the internal structure;

detecting voxels with contrast within the region of interest by applying a predetermined threshold;

performing connected components labeling to define a region of contrast;

computing area of defined region of contrast;

comparing area of defined region of contrast against a minimum size criteria; and

determining contrast is present within the internal structure when the defined area of contrast is greater than the minimum size criteria.

12. The method of claim 11 , wherein the internal structure is a heart, lung or colon.

13. The method of claim 11 , additionally comprising determining whether the internal structure has an elevated risk of irregularity based on the contrast.

14. The method of claim 11 , wherein the elevated risk of irregularity is a suspected emboli or cancer.

15. A computer system comprising:

a processor; and

a program storage device readable by the computer system, embodying a program of instructions executable by the processor to perform method steps for recognizing an internal structure from a three-dimensional image comprising a plurality of image sub-volumes the method comprising:

searching for an image sub-volume of the plurality of image sub-volume comprising a cross section of the internal structure using a binary search pattern to determine an order in which each of the plurality of image sub-volumes is analyzed; and

identifying bounds of the internal structure based on the location of the image sub-volume within which the internal structure is found.

16. The computer system of claim 15 , additionally comprising analyzing the three-dimensional image within the identified bounds of the internal structure to determine whether the internal structure has an elevated risk of irregularity.

17. The computer system of claim 16 , additionally comprising analyzing the three-dimensional image within the identified bounds of the internal structure to determine whether a contrast is present within the internal structure.

18. The computer system of claim 16 , wherein the image sub-volumes are image slices.