Systems and methods for automatic change quantification for medical decision support

1. A method of automatic change quantification for medical decision support, comprising:

automatically detecting a structure in a set of medical images;

characterizing the detected structure including modeling of deformation characteristics of the detected structure;

matching images based on the characterization of the detected structure, wherein a size measure of the detected structure is constrained according to the deformation characteristics; and

quantifying a change in the detected structure;

wherein estimating a uncertainty in the quantified change in the detected structure comprises estimating a confidence interval of a physiological, pathological or anatomical parameter measured from the detected structure.

2. The method of claim 1 , wherein modeling of deformation characteristics of the detected structure comprises at least one of elasticity modeling of tissues or structures, modeling of soft tissue characteristics, modeling of fat deformation characteristics, modeling of rigidity of bone structures, or articulatory modeling of joints between bones or cartilages.

3. The method of claim 1 , wherein modeling of deformation characteristics of the detected structure comprises modeling of pathological tissue deformation characteristics.

4. The method of claim 1 , wherein the size measure of the detected structure comprises a volume, a diameter, a radius, a projection length, a one-dimensional measure, two-dimensional measure or three-dimensional measure of the detected structure.

5. The method of claim 1 , wherein quantifying a change in the detected structure comprises quantifying a change in a physiological, pathological or anatomical parameter measured from the detected structure.

6. The method of claim 1 , further comprising estimating uncertainty in the quantified change in the detected structure.

7. The method of claim 6 , wherein estimating uncertainty in the quantified change in the detected structure comprises at least one of estimating process-induced uncertainties or estimating intrinsic structural uncertainties associated with the set of medical images.

8. The method of claim 6 , wherein estimating uncertainty in the quantified change in the detected structure comprises estimating tissue-dependent uncertainties.

9. The method of claim 6 , wherein the estimated uncertainty is fed back to constrain the detecting step.

10. The method of claim 6 , wherein the step of matching images includes a registration process, and wherein the estimated uncertainty is fed back to constrain the registration process.

11. The method of claim 6 , wherein the estimated uncertainty is fed back to constrain the step of quantifying a change in the detected structure.

12. A system for providing automatic change quantification for medical decision support, comprising:

a memory device for storing a program;

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

automatically detect a structure in a set of medical images;

characterize the detected structure including modeling of deformation characteristics of the detected structure;

match images based on the characterization of the detected structure, wherein a size measure of the detected structure is constrained according to the deformation characteristics; and

quantify a change in the detected structure;

wherein when estimating uncertainty in the quantified change in the detected structure, the processor is further operative with the program to estimate a change of a physiological, pathological or anatomical parameter measured from the detected structure.

13. The system of claim 12 , wherein when modeling deformation characteristics of the detected structure, the processor is further operative with the program to perform at least one of elasticity modeling of tissues or structures, modeling of soft tissue characteristics, modeling of fat deformation characteristics, modeling of rigidity of bone structures, or articulatory modeling of joints between bones or cartilages.

14. The system of claim 12 , wherein when modeling deformation characteristics of the detected structure, the processor is further operative with the program to perform modeling of deformation characteristics.

15. The system of claim 12 , wherein when matching images based on the characterization of the detected structure, the processor is further operative with the program to use the characterization of the structure to resolve matching ambiguities.

16. The system of claim 12 , wherein when quantifying a change in the detected structure, the processor is further operative with the program to quantify a change in a physiological, pathological or anatomical parameter measured from the detected structure.

17. The system of claim 12 , further comprising estimating uncertainty in the quantified change in the detected structure.

18. The system of claim 17 , wherein when estimating uncertainty in the quantified change in the detected structure, the processor is further operative with the program to estimate at least one of process-induced uncertainties or intrinsic structural uncertainties associated with the set of medical images.

19. The system of claim 17 , wherein when estimating uncertainty in the quantified change in the detected structure, the processor is further operative with the program to estimate tissue-dependent uncertainties.

20. The system of claim 17 , wherein the estimated uncertainty is fed back to constrain the detecting step.

21. The system of claim 17 , wherein when matching images based on the characterization of the detected structure, the processor is further operative with the program to perform a registration process, and wherein the estimated uncertainty is fed back to constrain the registration process.

22. A method of automatic tumor therapy monitoring, comprising:

automatically segmenting a tumor in a set of medical images using deformation characteristics of the tumor, wherein a size measure of the tumor is constrained according to the deformation characteristics during warping;

quantifying a change in a pathological parameter measured from the tumor;

and estimating uncertainty in the quantified change in the tumor;

wherein the estimated uncertainty is fed back to constrain the segmenting step.

23. The method of claim 22 , wherein estimating uncertainty in the quantified change in the tumor is based on patient clinical history or genetic background.

24. The method of claim 22 , further comprising modeling of deformation characteristics of the tumor.