Organ localization in biomedical image data using gradient fields cross-correlation
View Patent ↗This invention provides a system, method and computer-readable medium for detecting and localizing organs and other regions of interest in medical image data provided by a medical imaging procedure using gradient template information with respect to an example of the imaged organ and cross-correlation techniques to generate object detection information. In an embodiment, the detection and localization process/processor receives a plurality of gradient templates and employers a template with the closest matching cross-correlation to the gradient of the organ in the medical image.
1. A system for localizing a region of interest in a search image of at least a portion of a body comprising:
a gradient template generated prior to a runtime based upon an expected image property of the region of interest; and
a localization process/processor that, based upon the gradient template and during the runtime, identifies the region of interest and provides search image data that identifies portions of the region of interest;
wherein the localization process/processor is based upon a cross-correlation of normalized gradient fields corresponding to the search image of the body and the gradient template.
2. The system as set forth in claim 1 wherein the gradient template is based upon image properties in the region of interest within established boundaries around the region of interest.
3. The system as set forth in claim 2 wherein the region of interest is an organ.
4. The system as set forth in claim 3 wherein the organ is a prostate.
5. The system as set forth in claim 1 wherein the medical image includes image data acquired from a magnetic resonance imaging (MRI) device that images the body.
6. The system as set forth in claim 5 wherein the region of interest is an organ.
7. The system as set forth in claim 1 wherein the localization process/processor includes a segmenting process/processor that segments a representation of a boundary of the region of interest in response to results of a cross correlation computation.
8. The system as set forth in claim 1 wherein the medical image is characterized by intensity invariance.
9. The system as set forth in claim 8 wherein the medical image comprises a MRI-based image.
10. The system as set forth in claim 1 wherein the region of interest defines a pelvic region and a prostate.
11. The system as set forth in claim 1 wherein the gradient template includes a plurality of template components defined in three dimensions.
12. The system as set forth in claim 11 wherein the plurality of template components include an x-dimension template component, a y-dimension template component and a z-dimension template component.
13. The system as set forth in claim 1 wherein the gradient template includes at least one of (a) a plurality of boundary gradients with respect to an organ and (b) boundary gradients that enclose an organ boundary.
14. The system as set forth in claim 1 wherein the gradient template is constructed and arranged to characterize an organ at a plurality of discrete scales.
15. The system as set forth in claim 14 wherein the discrete scales include discrete aspect ratios.
16. The system as set forth in claim 1 further comprising an output process/processor that outputs a visual representation of a detected and localized organ to a display.
17. A method for localizing a region of interest in a search image of at least a portion of a body comprising the steps of:
providing, prior to a runtime, a stored gradient template based upon an expected image property of the region of interest; and
detecting and localizing, based upon the gradient template and during the runtime, the region of interest and provides search image data that identifies portions of the region of interest;
wherein the step of detecting and localizing includes cross-correlating normalized gradient fields corresponding to the search image of the body and the gradient template.
18. The method as set forth in claim 17 wherein the region of interest is an organ.
19. The method as set forth in claim 18 wherein the organ is a prostate.
20. The method as set forth in claim 17 wherein the medical image includes image data acquired from a magnetic resonance imaging (MRI) device that images the body.
21. The method as set forth in claim 1 further comprising segmenting a representation of a boundary of the region of interest in response to results of a cross correlation computation.
22. The method as set forth in claim 17 wherein the medical image is characterized by intensity invariance.
23. The method as set forth in claim 22 wherein the medical image comprises a MRI-based image.
24. The method as set forth in claim 17 wherein the region of interest defines a pelvic region and a prostate.
25. The method as set forth in claim 17 wherein the step of providing the gradient template includes defining a plurality of template components defined in three dimensions.
26. The method as set forth in claim 25 wherein the plurality of template components include an x-dimension template component, a y-dimension template component and a z-dimension template component.
27. The method as set forth in claim 17 wherein the gradient template includes at least one of (a) a plurality of boundary gradients with respect to an organ and (b) boundary gradients that enclose an organ boundary.
28. The method as set forth in claim 17 wherein the gradient template characterizes an organ at a plurality of discrete scales.
29. The method as set forth in claim 28 wherein the discrete scales include discrete aspect ratios.
30. The method as set forth in claim 17 further comprising outputting a visual representation of a detected and localized organ to a display.
31. A non-transitory computer-readable medium including program instructions comprising:
a data structure that includes one or more image gradient templates generated prior to a runtime with respect to at least one organ that is subjected to medical imaging; and
an image-processing procedure that, during the runtime, employs the data structure to derive information about the at least one organ from image data by cross-correlating normalized gradient fields corresponding to a search image of the at least one organ and the one or more image gradient templates.
32. The computer-readable medium as set forth in claim 31 wherein the image-processing procedure is constructed and arranged to detect and localize the organ in the image data.
33. The computer-readable medium as set forth in claim 32 wherein the image-processing procedure is constructed and arranged to segment a representation of a boundary of the organ based upon results of cross-correlating.
34. A method for localizing an organ in a medical image of at least a portion of a body comprising the steps of:
providing a plurality of stored gradient templates based upon an organ prior to a runtime;
receiving the medical image; and
detecting the organ in the medical image, during the runtime, based upon at least one of the gradient templates in the plurality of gradient templates that produces a maximum cross-correlation with normalized gradients in the medical image.
35. The method as set forth in claim 34 wherein the organ comprises a prostate.
36. The method as set forth in claim 34 further comprising outputting a visual representation of a detected organ to a display.
37. The system as set forth in claim wherein the gradient template is normalized.