Method and system for advanced aneurysm analysis

1. An automated method for aneurysm analysis, comprising:

extracting shape descriptors from test vessel data;

generating an aneurysm probability map for the test vessel data using the shape descriptors;

detecting the presence of the aneurysm on the test vessel data;

localizing the aneurysm in the probability map; and

separating the aneurysm from the probability map,

wherein separating the aneurysm from the probability map comprises:

extracting shape descriptors that can identify the aneurysm from a database;

computing a weak unary potential based on the probability map and a strong unary potential based on the intersection of the test vessel data with three planes defined by a dome and proximal and distal points of the aneurysm;

computing pairwise potentials;

combining the weak and strong unary potentials with the pairwise potentials to formulate an energy function that depends on label assignment in the probability map; and

optimizing the energy function.

2. The method of claim 1 , wherein the shape descriptors are extracted to capture local geometric and regional shape information of the aneurysm.

3. The method of claim 2 , wherein the probability map is generated by using learned properties of the extracted information.

4. The method of claim 1 , wherein the presence of the aneurysm is detected by finding if the probability map has a value greater than a minimum probability threshold.

5. The method of claim 1 , wherein the aneurysm is localized by determining the most probable aneurysm location from the probability map if the aneurysm is detected.

6. The method of claim 1 , wherein the aneurysm is separated from the probability map by executing a separation algorithm on a region of interest including the aneurysm.

7. The method of claim 6 , wherein the separation algorithm automatically finds the region of interest, and the proximal and distal points.

8. The method of claim 1 , further comprising quantifying the aneurysm.

9. The method of claim 1 , further comprising planning a surgical procedure by using the aneurysm.

10. The method of claim 1 , further comprising:

localizing another aneurysm in the probability map; and

separating the another aneurysm from the probability map.

11. A system for automated aneurysm analysis, comprising:

a memory device for storing a program;

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

extract shape descriptors from test vessel data;

generate an aneurysm probability map for the test vessel data using the shape descriptors;

detect the presence of the aneurysm on the test vessel data;

localize the aneurysm in the probability map; and

separate the aneurysm from the probability map,

wherein the processor is further operative with the program when separating the aneurysm from the probability map to:

extract shape descriptors that can identify the aneurysm from a database;

compute a weak unary potential based on the probability map and a strong unary potential based on the intersection of the test vessel data with three planes defined by a dome and proximal and distal points of the aneurysm;

compute pairwise potentials;

combine the weak and strong unary potentials with the pairwise potentials to formulate an energy function that depends on label assignment in the probability map; and

optimize the energy function.

12. The system of claim 11 , wherein the shape descriptors are extracted to capture local geometric and regional shape information of the aneurysm.

13. The system of claim 12 , wherein the probability map is generated by using learned properties of the extracted information.

14. The system of claim 11 , wherein the presence of the aneurysm is detected by finding if the probability map has a value greater than a minimum probability threshold.

15. The system of claim 11 , wherein the aneurysm is localized by determining the most probable aneurysm location from the probability map if the aneurysm is detected.

16. The system of claim 11 , wherein the aneurysm is separated from the probability map by executing a separation algorithm on a region of interest including the aneurysm.

17. The system of claim 16 , wherein the separation algorithm automatically finds the region of interest, and the proximal and distal points.

18. The system of claim 11 , wherein the processor is further operative with the program to quantify the aneurysm.

19. The system of claim 11 , wherein the processor is further operative with the program to plan a surgical procedure by using the aneurysm.

20. The system of claim 11 , wherein the processor is further operative with the program to:

localize another aneurysm in the probability map; and

separate the another aneurysm from the probability map.