Generating applications that analyze image data using a semantic cognition network

• US 7,467,159 B2
• Filed: 10/17/2006
• Issued: 12/16/2008
• Est. Priority Date: 10/15/2002
• Status: Active Grant

• Current Assignee: Definiens AG (Astrazeneca PLC)
• Original Assignee: Definiens AG (Astrazeneca PLC)
• Inventors: Binnig, Gerd, Athelogou, Maria, Krug, Christof, Schaepe, Arno, Benz, Ursula
• Primary Examiner(s): LEWIS, CHERYL RENEA
• Application Number:
  US11/582,225
• Publication Number:
  US 20070036440A1
• Time in Patent Office: 791 Days
• Field of Search: 707103 R-103 Z, 707/102, 707/104.1, 717/100, 717/101, 717/120, 717/121
• US Class Current: 1/1
• CPC Class Codes: G06N 5/022 Knowledge engineering; Knowledge acquisition
• CPC Class Codes: G06V 10/248 by interactive preprocessing or interactive shape modelling, e.g. feature points assigned by a user
• CPC Class Codes: G06V 10/422 for representing the structure of the pattern or shape of an object therefor
• CPC Class Codes: G06V 20/69 Microscopic objects, e.g. biological cells or cellular parts
• CPC Class Codes: Y10S 707/99943 Generating database or data structure, e.g. via user interface
• CPC Class Codes: Y10S 707/99944 Object-oriented database structure
• CPC Class Codes: Y10S 707/99945 Object-oriented database structure processing
• CPC Class Codes: Y10S 707/99946 Object-oriented database structure network

Abstract

A method is disclosed for generating an application that analyzes image data, such as from satellite and microscope pictures. The method uses a graphical user interface to add a new processing object to a processing object network. The processing object network includes a parent processing object and a child processing object. A user can append a new processing object to the child processing object or can add the new processing object as a subprocess to the parent processing object. The user selects a data domain and an algorithm from selection lists on the graphical user interface and adds them to the new processing object. The application uses a semantic cognition network to process data objects that are generated by segmenting the image data. The application then uses the new processing object to identify portions of the image that are to be highlighted on the graphical user interface.

Claims

1. A system for generating an application that analyzes image information, comprising:

a computer readable storage medium; and

a computer program stored on the storage medium comprising;

a set of algorithms, wherein each algorithm of the set of algorithms represents an operation performable by the application;

a plurality of data domains, wherein each of the plurality of data domains represents a subset of a data object network, and wherein the data object network is generated by segmenting the image information; and

a graphical user interface usable to generate a processing object network, wherein the processing object network includes a parent process and a plurality of child processes, and wherein the parent process comprises one of the plurality of data domains and one algorithm of the set of algorithms.

2. The system of claim 1, wherein the computer program displays on the graphical user interface a list of each of the plurality of data domains and a list of each algorithm of the set of algorithms.

3. The system of claim 1, wherein a digital image comprises the image information, and wherein the computer program displays on the graphical user interface the data object network as an overlay over the digital image.

4. The system of claim 1, wherein the graphical user interface includes a dialog element in which a selection list of the set of algorithms is displayed.

5. The system of claim 1, wherein the graphical user interface includes a dialog element in which a selection list of the plurality of data domains is displayed.

6. The system of claim 1, wherein the parent process and the plurality of child processes are linked with each other to form an execution control for the application.

7. The system of claim 1, wherein the processing object network comprises a control structure of the application, and wherein the control structure is taken from the group consisting of a loop, a branch, a function call, a jump and an execution control.

8. The system of claim 1, wherein the plurality of child processes are linked with each other, and wherein the application executes one of the plurality of child processes as a sub-process.

9. The system of claim 1, wherein the subset of the data object network comprises data objects, and wherein the application executes the parent process several times, each time processing additional data objects.

10. The system of claim 1, wherein the computer program generates a semantic cognition network, and wherein the semantic cognition network comprises the data object network, the processing object network, and a class object network.

11. The system of claim 10, wherein the class object network comprises a class domain, and wherein the parent process is unambiguously defined by the one of the plurality of data domains, the one algorithm of the set of algorithms, and the class domain.

12. The system of claim 10, wherein the application performs a morphological operation using the one algorithm of the set of algorithms to combine data objects of the subset of the data object network with additional data objects to create secondary data objects.

13. The system of claim 12, wherein the application compares the secondary data objects with the class domain.

14. The system of claim 1, wherein the data object network and the processing object network are linked together by link objects that are dynamically generated by the application based on the one of the plurality of data domains.

15. The system of claim 1, wherein the parent process extracts a subset of data objects from the data object network and applies the one algorithm of the set of algorithms to each data object in the subset of data objects.

16. A system for generating an application that analyzes digital image data, comprising:

a computer readable storage medium; and

a computer program stored on the storage medium comprising;

a set of algorithms, wherein each algorithm of the set of algorithms represents an operation performable by the application;

a plurality of data domains, wherein each of the plurality of data domains represents a subset of a data object network, and wherein the data object network is generated by segmenting the digital image data; and

means for generating a processing object network, wherein the processing object network includes a parent process and a plurality of child processes, and wherein the parent process comprises one of the plurality of data domains and one algorithm of the set of algorithms.

17. The system of claim 16, wherein the processing object network comprises a control structure of the application, and wherein the control structure is taken from the group consisting of a loop, a branch, a function call, a jump and an execution control routine.

18. The system of claim 16, wherein the subset of the data object network comprises data objects, and wherein the application executes the parent process several times, each time processing additional data objects.

19. The system of claim 16, wherein the parent process extracts a subset of data objects from the data object network and applies the one algorithm of the set of algorithms to each data object in the subset of data objects.

20. The system of claim 16, wherein the computer program displays a selection list of available data domains, and wherein the means generates the processing object network in response to the one of the plurality of data domains being selected from the selection list of available data domains.

21. A computer implemented method for generating an application that analyzes an image, comprising:

adding a new processing object to a processing object network using a graphical user interface, wherein the processing object network includes a parent processing object and a child processing object, and wherein the image is displayed on the graphical user interface;

designating that the new processing object is appended to the child processing object;

selecting a data domain from a list of data domains displayed in a first dialog element on the graphical user interface and adding the data domain to the new processing object; and

selecting an algorithm from a list of algorithms displayed in a second dialog element on the graphical user interface and adding the algorithm to the new processing object, wherein the application uses the new processing object to identify a portion of the image that is to be highlighted on the graphical user interface.

22. The method of claim 21, further comprising:

configuring parameters of the algorithm by selecting from among available parameters displayed in a selection list in a third dialog element.

23. The method of claim 21, wherein the designating that the new processing object is appended is performed by selecting from a selection list in a third dialog element, and wherein the selection list includes an append option and a create subprocess option.

24. The method of claim 21, wherein the selecting the data domain is performed by clicking on an option listed in a selection list of available data domains.