TEMPLATE-LESS METHOD FOR ARBITRARY RADIOPAQUE OBJECT TRACKING IN DYNAMIC IMAGING

• US 20150254859A1
• Filed: 07/31/2014
• Published: 09/10/2015
• Est. Priority Date: 07/31/2013
• Status: Active Grant

• Current Assignee: Varian Medical Systems Incorporated (Siemens AG)
• Original Assignee: Materials Engineering And Packaging, LLC
• Inventors: Parikh, Parag J., Wan, Hanlin
• Granted Patent: US 9,471,985 B2
• Time in Patent Office: Days
• US Class Current: 1/1
• CPC Class Codes: A61N 5/1049 for verifying the position of the patient with respect to the radiation beam
• CPC Class Codes: G06T 2207/10081 Computed x-ray tomography [CT]
• CPC Class Codes: G06T 2207/10116 X-ray image
• CPC Class Codes: G06T 2207/30004 Biomedical image processing
• CPC Class Codes: G06T 2207/30204 Marker
• CPC Class Codes: G06T 7/74 involving reference images or patches

Abstract

A method for localizing a fiducial marker in a medical image includes: obtaining the medical image; and performing an analysis, using an analysis unit, to localize the fiducial marker in the medical image; wherein the fiducial marker is localized by the analysis unit without use of a marker template.

Claims

1. A method for localizing a fiducial marker in a medical image comprising:

obtaining the medical image; and

performing an analysis, using an analysis unit, to localize the fiducial marker in the medical image;

wherein the fiducial marker is localized by the analysis unit without use of a marker template.

2. The method of claim 1, further comprising determining a search window of the medical image.

3. The method of claim 2, wherein the search window is determined based on a source-object-imager geometry.

4. The method of claim 2, wherein the search window is determined based on a previously obtained analysis result.

5. The method of claim 2, wherein the act of performing the analysis comprises identifying one or more minima in the search window.

6. The method of claim 5, wherein the one or more minima in the search window are identified based on an intensity contrast with a surrounding background intensity.

7. The method of claim 5, wherein the act of performing the analysis further comprises determining a cost function.

8. The method of claim 7, wherein the act of performing the analysis further comprises minimizing the cost function.

9. The method of claim 7, wherein the cost function comprises a compound cost function.

10. The method of claim 7, wherein the cost function includes a component representing a rescaled grayscale intensity at a minima in the search window.

11. The method of claim 7, wherein the cost function includes a component representing a distance from a reference location in the search window to a minima in the search window.

12. The method of claim 7, wherein the cost function includes a component representing a distance between a minima in the search window of the medical image and a minima in another search window of a successive medical image.

13. The method of claim 7, wherein the cost function includes a first component representing a rescaled grayscale intensity at a minima in the search window, a second component representing a distance from a reference location in the search window to the minima in the search window, and a third component representing a distance between the minima in the search window of the medical image and a minima in another search window of a successive medical image.

14. The method of claim 8, wherein the act of minimizing the cost function results in a determination of a marker position for the fiducial marker.

15. The method of claim 14, further comprising repeating the act of performing the analysis to determine an additional marker position for the fiducial marker.

16. The method of claim 15, further comprising comparing the marker position with the additional marker position.

17. The method of claim 2, wherein the act of performing the analysis comprises identifying one or more maxima in the search window.

18. The method of claim 14, wherein the act of performing the analysis further comprises determining a cost function and maximizing the cost function.

19. An apparatus for localizing a fiducial marker in a medical image comprising:

a processing unit configured to obtain the medical image, and performing an analysis to localize the fiducial marker in the medical image, wherein the processing unit is configured to localize the fiducial marker without use of a marker template; and

a medium for storing a position of the fiducial marker.

20. The apparatus of claim 19, wherein the processing unit is further configured to determine a search window of the medical image.

21. The apparatus of claim 20, wherein the processing unit is configured to perform the analysis by identifying one or more minima in the search window.

22. The apparatus of claim 21, wherein the processing unit is configured to identify one or more minima in the search window based on an intensity contrast with a surrounding background intensity.

23. The apparatus of claim 21, wherein the processing unit is configured to perform the analysis further by determining a cost function.

24. The apparatus of claim 23, wherein the processing unit is configured to perform the analysis further by minimizing the cost function.

25. The apparatus of claim 23, wherein the cost function comprises a compound cost function.

26. The apparatus of claim 23, wherein the cost function includes a component representing a rescaled grayscale intensity at a minima in the search window.

27. The apparatus of claim 23, wherein the cost function includes a component representing a distance from a reference location in the search window to a minima in the search window.

28. The apparatus of claim 23, wherein the cost function includes a component representing a distance between a minima in the search window of the medical image and a minima in another search window of a successive medical image.

29. The apparatus of claim 23, wherein the cost function includes a first component representing a rescaled grayscale intensity at a minima in the search window, a second component representing a distance from a reference location in the search window to the minima in the search window, and a third component representing a distance between the minima in the search window of the medical image and a minima in another search window of a successive medical image.

30. The apparatus of claim 24, wherein the act of minimizing the cost function results in a determination of a marker position for the marker.

31. The apparatus of claim 30, wherein the processing unit is further configured to repeat the analysis to determine an additional marker position for the fiducial marker.

32. The apparatus of claim 31, wherein the processing unit is further configured to compare the marker position with the additional marker position.

33. The apparatus of claim 20, wherein the processing unit is configured to perform the analysis by identifying one or more maxima in the search window.

34. The apparatus of claim 33, wherein the processing unit is configured to perform the analysis further by determining a cost function and maximizing the cost function.

35. A method for image analysis comprising:

I) manipulating a set of images to generate at least one approximated search window in each image such that the at least one approximated search window is located in the vicinity of a point of interest in the images;

II) performing a first dynamic analysis, relative to the approximated search windows by;

A) identifying within the at least one approximated search window generated in each image one or more contrast points, a contrast point being a local point having an imaging intensity presenting a peak contrast relative to the imaging intensities of points surrounding the local point;

B) performing a contrast point characteristic identification step of determining, for each contrast point identified in an individual approximated search window in an individual image of the set of images, at least one of the following contrast point characteristics;

i) a distance value between the contrast point and a reference point of the individual approximated search window;

ii) an imaging intensity value of the contrast point; and

iii) a plurality of distance values measuring the distance from the location of the contrast point in the individual approximated search window to the location of each contrast point in the at least one approximated search window in another image of the set of images;

C) repeating the contrast point characteristic identification step for the at least one approximated search window in each image of the image set;

D) executing a dynamic programming algorithm to calculate a first resultant point for an individual approximated search window in an individual image of the image set, wherein the dynamic programming algorithm is executed based, at least in part, on input values corresponding to at least one contrast point characteristic value determined during the contrast point characteristic identification step, with an input value being provided for each contrast point identified in the individual approximated search window;

E) repeating execution of the dynamic programming algorithm for the at least one approximated search window in each image of the image set; and

F) storing the calculated first resultant point of each approximate search window in a data storage unit;

III) manipulating the set of images to generate at least one refined search window in each image such that the at least one refined search window is positioned based on the first resultant point calculated relative to the individual approximated search window for that respective image during the first dynamic analysis;

IV) performing a second dynamic analysis, relative to the refined search windows by repeating the steps of the first dynamic analysis (II), though performing the repeated steps in the second dynamic analysis relative to the refined search windows to calculate and store a second resultant point for each refined search window; and

V) performing a convergence judgment step to judge if a convergence has occurred in the image analysis by;

G) performing a comparison step of comparing the first resultant point calculated for the at least one approximate search window generated an individual image of the image set with the second resultant point calculated for the at least one refined search window generated in the same individual image to determine if the first resultant point and the second result point are identical;

H) repeating the comparison step for the first resultant point and the second resultant point calculated for each image of the image set;

I) calculating a convergence percentage by calculating the percent of images of the image set that have a first resultant point and a second resultant point that are identical; and

J) judging whether the calculated convergence percentage satisfies a predetermined convergence threshold, with a convergence being deemed to have occurred if the calculated convergence percentage is equal to or above the predetermined convergence threshold, and a convergence being deemed to not have occurred if the calculated convergence percentage is below the predetermined convergence threshold,

wherein, if it is deemed in the convergence judgment step (J) that a convergence has occurred, then the identical point identified by the first resultant point and the second resultant point in each image of the image set is deemed to be the point of interest, and

wherein, if it is deemed in the convergence judgment step (J) that a convergence has not occurred, then steps (III), (IV) and (V) are repeated to calculate and store a third resultant point for each further refined search window, and to judge convergence in the image analysis by comparing the third resultant points to the second resultant points.

36. The method for image analysis according to claim 35, wherein the contrast point is a local minimum point, being a local point having a minimum imaging intensity relative to the imaging intensities of points surrounding the local point.

37. The method for image analysis according to claim 35, wherein the contrast point is a local maximum point, being a local point having a maximum imaging intensity relative to the imaging intensities of points surrounding the local point.