Intelligent tumor tracking system

1. A method comprising:

receiving a candidate segmentation of a current medical image, the candidate segmentation representing an estimated boundary of an anatomical feature in the current medical image;

providing the candidate segmentation to a first trained classifier; and

receiving from the first trained classifier a first indication of accuracy of the candidate segmentation based on one or more feature of the candidate segmentation, the first indication of accuracy relating to at least one of comparison of the estimated boundary to edges in the medical image, a determination of smoothness of the estimated boundary, or a determination of continuity of the estimated boundary;

receiving one or more prior segmentation of a prior medical image of the anatomical feature, wherein said one or more prior segmentation does not include the candidate segmentation;

providing the candidate segmentation and said one or more prior segmentation to a second trained classifier;

receiving from the second trained classifier a second indication of accuracy of the candidate segmentation based on said one or more prior segmentation.

2. The method of claim 1 , wherein the first indication of accuracy of the candidate segmentation comprises a probability.

3. The method of claim 1 , wherein the first trained classifier comprises an artificial neural network.

4. The method of claim 1 , wherein the second trained classifier comprises an artificial neural network.

5. The method of claim 1 , further comprising: displaying to a user the candidate segmentation and said one or more prior segmentation;

prompting the user to adjust or approve the candidate segmentation.

6. The method of claim 1 , further comprising:

displaying to a user the candidate segmentation;

prompting the user to adjust the candidate segmentation when the first indication of accuracy fails to meet a predetermined threshold.

7. The method of claim 1 , wherein the candidate segmentation is automatically generated.

8. The method of claim 1 , further comprising:

training the first trained classifier based on a plurality of automatically generated segmentations.

9. A system comprising:

a computing node comprising a computer readable storage medium having program instructions embodied therewith, the program instructions executable by a processor of the computing node to cause the processor to perform a method comprising:

receiving a candidate segmentation of a current medical image, the candidate segmentation representing an estimated boundary of an anatomical feature in the current medical image;

providing the candidate segmentation to a first trained classifier;

receiving from the first trained classifier a first indication of accuracy of the candidate segmentation based on one or more feature of the candidate segmentation, the first indication of accuracy relating to at least one of comparison of the estimated boundary to edges in the medical image, a determination of smoothness of the estimated boundary, or a determination of continuity of the estimated boundary;

receiving one or more prior segmentation of a prior medical image of the anatomical feature, wherein said one or more prior segmentation does not include the candidate segmentation;

providing the candidate segmentation and said one or more prior segmentation to a second trained classifier; and

receiving from the second trained classifier a second indication of accuracy of the candidate segmentation based on said one or more prior segmentation.

10. A computer program product for evaluation of segmentation of medical imagery, the computer program product comprising a computer readable storage medium having program instructions embodied therewith, the program instructions executable by a processor to cause the processor to perform a method comprising:

receiving a candidate segmentation of a current medical image, the candidate segmentation representing an estimated boundary of an anatomical feature in the current medical image;

providing the candidate segmentation to a first trained classifier;

receiving from the first trained classifier a first indication of accuracy of the candidate segmentation based on one or more feature of the candidate segmentation, the first indication of accuracy relating to at least one of comparison of the estimated boundary to edges in the medical image, a determination of smoothness of the estimated boundary, or a determination of continuity of the estimated boundary;

receiving one or more prior segmentation of a prior medical image of the anatomical feature, wherein said one or more prior segmentation does not include the candidate segmentation;

providing the candidate segmentation and said one or more prior segmentation to a second trained classifier;

receiving from the second trained classifier a second indication of accuracy of the candidate segmentation based on said one or more prior segmentation.

11. The computer program product of claim 10 , wherein the first indication of accuracy of the candidate segmentation comprises a probability.

12. The computer program product of claim 10 , wherein the first trained classifier comprises an artificial neural network.

13. The computer program product of claim 10 , wherein the second trained classifier comprises an artificial neural network.

14. The computer program product of claim 10 , the method further comprising:

displaying to a user the candidate segmentation and said one or more prior segmentation;

prompting the user to adjust or approve the candidate segmentation.

15. The computer program product of claim 10 , the method further comprising:

displaying to a user the candidate segmentation;

prompting the user to adjust the candidate segmentation when the first indication of accuracy fails to meet a predetermined threshold.

16. The computer program product of claim 10 , the method further comprising:

training the first trained classifier based on a plurality of automatically generated segmentations.