System and method for controlling user repeatability and reproducibility of automated image annotation correction
Systems and methods are disclosed for controlling image annotation. One method includes acquiring a digital representation of image data and generating a set of image annotations for the digital representation of the image data. The method also may include determining an association between members of the set of image annotations and generating one or more groups of members based on the association. A representative annotation from the one or more groups may also be determined, presented for selection, and the selection may be recorded in memory.
1. A computer-implemented method of controlling image annotation, using at least one computer system, the method comprising using a processor for:
receiving a digital representation of image data corresponding to anatomic features of a patient;
segmenting the digital representation to determine a first candidate coronary vessel tree by a method controlled by one or more parameters;
segmenting the digital representation to generate a plurality of alternative candidate coronary vessel trees corresponding to the first candidate coronary vessel tree:
generating a plurality of groups of alternative candidate coronary vessel trees, each group of the plurality of groups comprising a subset of the plurality of alternative candidate coronary vessel trees;
determining, by the at least one computer system, a representative alternative candidate coronary vessel tree for each of the groups of alternative candidate coronary vessel trees, the representative alternative candidate coronary vessel tree for each group being determined as either:
an alternative candidate coronary vessel tree of the group closest to an average alternative candidate coronary vessel tree of the group, or
a generated average coronary vessel tree of the group; and
presenting, by the at least one computer system, the first candidate coronary vessel tree and each representative alternative candidate coronary vessel tree for selection and/or correction by a user.
2. The computer-implemented method of claim 1 ,
wherein the plurality of alternative candidate coronary vessel trees are generated by one or more of randomly selecting settings of the one or more parameters, applying a plurality of image segmentation techniques, perturbing the digital representation by rigid transformation, or perturbing the digital representation by an addition of Gaussian noise to the digital representation.
3. The computer-implemented method of claim 1 , further comprising:
determining, by the at least one computer system, one or more associations between alternative candidate coronary vessel trees of the plurality of alternative candidate coronary vessel trees,
wherein the plurality of groups of alternative candidate coronary vessel trees are generated based on the determined one or more associations.
4. The computer-implemented method of claim 1 , wherein the digital representation of image data is acquired electronically via a network.
5. The computer-implemented method of claim 1 , wherein each representative coronary vessel tree identifies anatomic features identified within the digital representation of image data.
6. The computer-implemented method of claim 1 , wherein the generated first candidate coronary vessel tree is generated using one or more of a level set, a graph based method, an active shape model, an active appearance model.
7. The computer-implemented method of claim 3 , wherein each of the one or more associations between alternative candidate coronary vessel trees of the plurality of alternative candidate coronary vessel trees is determined by computing a dice coefficient.
8. The computer-implemented method of claim 1 , wherein generating the plurality of groups of alternative candidate coronary vessel trees includes generating the plurality of groups of alternative candidate coronary vessel trees via a cluster algorithm.
9. The computer-implemented method of claim 8 , wherein the cluster algorithm includes spectral partitioning and isoperimetric portioning.
10. The computer-implemented method of claim 1 , wherein the image data is a portion of a body organ.
11. A system of controlling image annotation, the system comprising:
a data storage device storing instructions for controlling image segmentation; and
a processor configured to execute the instruction to perform a method including:
receiving a digital representation of image data corresponding to anatomic features of a patient;
segmenting the digital representation to determine a first candidate coronary vessel tree by a method controlled by one or more parameters;
segmenting the digital representation to generate a plurality of alternative candidate coronary vessel trees corresponding to the first candidate coronary vessel tree:
generating a plurality of groups of alternative candidate coronary vessel trees, each group of the plurality of groups comprising a subset of the plurality of alternative candidate coronary vessel trees;
determining a representative alternative candidate coronary vessel tree for each of the groups of alternative candidate coronary vessel trees, the representative alternative candidate coronary vessel tree for each group being determined as either:
an alternative candidate coronary vessel tree of the group closest to an average alternative candidate coronary vessel tree of the group, or
a generated average coronary vessel tree of the group; and
presenting the first candidate coronary vessel tree and each representative alternative candidate coronary vessel tree for selection or correction by a user.
12. The system of claim 11 , wherein the plurality of alternative candidate coronary vessel trees are generated by one or more of randomly selecting settings of the one or more parameters, applying a plurality of image segmentation techniques, perturbing the digital representation by rigid transformation, or perturbing the digital representation by an addition of Gaussian noise to the digital representation.
13. The system of claim 11 , wherein the processor is further configured to electronically acquire the digital representation of image data via a network.
14. The system of claim 11 , wherein the processor is further configured to determine one or more associations between alternative candidate coronary vessel trees of the plurality of alternative candidate coronary vessel trees,
wherein the plurality of groups of alternative candidate coronary vessel trees are generated based on the determined one or more associations.
15. The system of claim 11 , wherein the processor is further configured to automatically generate the first candidate coronary vessel tree using one or more of a level set, a graph based method, an active shape model, an active appearance model.
16. The system of claim 14 , wherein the processor is configured to determine each of the one or more associations between alternative candidate coronary vessel trees of the plurality of alternative candidate of coronary vessel trees by computing a dice coefficient.
17. The system of claim 14 , wherein processor is configured to generate the plurality of groups of alternative candidate coronary vessels trees via a cluster algorithm.
18. The system of claim 17 , wherein the cluster algorithm includes spectral partitioning and isoperimetric portioning.
19. A non-transitory computer-readable medium for use on a computer system containing computer-executable programming instructions for controlling image annotation, the instructions being executable by the computer system for:
receiving a digital representation of image data corresponding to anatomic features of a patient;
segmenting the digital representation to determine a first candidate coronary vessel tree by a method controlled by one or more parameters;
segmenting the digital representation to generate a plurality of alternative candidate coronary vessel trees corresponding to the first candidate coronary vessel tree:
generating a plurality of groups of alternative candidate coronary vessel trees, each group of the plurality of groups comprising a subset of the plurality of alternative candidate coronary vessel trees;
determining, by the computer system, a representative alternative candidate coronary vessel tree for each of the groups of alternative candidate coronary vessel trees, the representative alternative candidate coronary vessel tree for each group being determined as either:
an alternative candidate coronary vessel tree of the group closest to an average alternative candidate coronary vessel tree of the group, or
a generated average coronary vessel tree of the group; and
presenting, by the computer system, the first candidate coronary vessel tree and each representative alternative candidate coronary vessel tree for selection or correction by a user.
20. The non-transitory computer readable medium of claim 19 , the instructions being executable by the computer system further comprising determining one or more associations between alternative candidate coronary vessel trees of the plurality of alternative candidate coronary vessel trees,
wherein the plurality of groups of alternative candidate coronary vessel trees are generated based on the determined one or more associations.