processing data for interpretation

• US 7,200,612 B2
• Filed: 03/23/2001
• Issued: 04/03/2007
• Est. Priority Date: 03/23/2000
• Status: Expired due to Term

• Current Assignee: Siemens Medical Solutions USA Incorporated (Siemens AG)
• Original Assignee: Mirada Solutions Ltd. (Siemens AG)
• Inventors: Brady, John Michael, Feldmar, Jacques
• Primary Examiner(s): Wong; Don
• Assistant Examiner(s): SHECHTMAN, CHERYL MARIA
• Application Number:
  US10/221,992
• Publication Number:
  US 20030144976A1
• Time in Patent Office: 2,202 Days
• Field of Search: 707/104.1, 707/10, 382/156, 382/128, 705/2, 705/3
• US Class Current: 1/1
• CPC Class Codes: G06N 5/04 Inference or reasoning models
• CPC Class Codes: G16H 10/60 for patient-specific data, e.g. for electronic patient records
• CPC Class Codes: G16H 50/50 for simulation or modelling of medical disorders
• CPC Class Codes: G16H 70/20 relating to practices or guidelines
• CPC Class Codes: Y10S 707/99945 Object-oriented database structure processing
• CPC Class Codes: Y10S 707/99948 Application of database or data structure, e.g. distributed, multimedia, or image

Abstract

A system for improving sensor-based decision making provides for the automatic submission of data obtained locally from instrumentation (such as image data) together with the interpretation of that data, which can be the output of some software which has been checked and possibly corrected by a user according to his/her expertise, to a remote database via an internetwork. The submission to the remote database is preferably automatic so that the remote database grows over time. The local site can access the remote database to retrieve information to assist in interpretation of the locally produced data (for example similar images and their corresponding interpretations), or can retrieve updated or improved software or parameters improving the software used for processing the data. The information on the remote database can also be reprocessed by software agents to provide statistical information based on information from a variety of such local sites. The system is particularly useful in improving the interpretation of data which is difficult to interpret such as medical image data (e.g. mammographic or cardiac ultrasound data).

Claims

1. A method of processing data for a user, comprising:

accepting input of data acquired from instrumentation as a dataset of an information domain which dataset comprises image data which comprises medical image data; and

processing the data set under the control of processing software, said processing comprising the input of “

user truth”

based on interpretation of the dataset by the user, to provide a processed dataset to assist in further interpretation of the dataset;

storing via a network said dataset and said processed dataset into a database of information relating to said domain acquired from a plurality of users;

estimating automatically and quantitatively the effect of said “

user truth”

on the data set-in order to combine in said database such data from multiple users while minimizing the effects of user truth;

and wherein said processing software comprises a software agent operative, on the basis of at least one of attributes of said dataset, attributes of the processed dataset and user-input, to access the database to retrieve data and/or software to mediate the processing and/or interpretation of said dataset by said processing software wherein the software agent is operative to transfer to the database data concerning manual adjustments made to said processed dataset by the user, and constituting said “

user truth”

, wherein the processing of said dataset is mediated by updating said processing software using at least one of data and software from the database.

2. A method according to claim 1 wherein the storing via the network of said dataset and said processed dataset into the database is automatic.

3. A method according to claim 1 wherein the software agent comprises means for assisting the user to prepare the dataset and, processed dataset for storing in the database.

4. A method according to claim 1, wherein said database comprises information produced by execution of said method by said plurality of users at a plurality of geographically separate sites.

5. A method according to claim 4, wherein said information produced by execution of said method comprises datasets and processed datasets from said plurality of geographically separate sites.

6. A method according to claim 5, further comprising comparing together the datasets from said plurality of geographically separate sites stored on the database.

7. A method according to claim 6, further comprising storing the results of comparing together the datasets from said plurality of geographically separate sites stored on the database, and outputting the results to the plurality of geographically separate sites as said data to mediate the processing or interpretation of said dataset.

8. A method according to claim 6 wherein said datasets are processed under the control of different processing software at different ones of said geographically separate sites.

9. A method according to claim 5 further comprising processing said information in the database to generate data representing at least one of:

the quality of, average values of, and principle components of said datasets or processed datasets.

10. A method according to claim 5 further comprising validating the information in said database.

11. A method according to claim 10 wherein the validation is by cross-checking between different users.

12. A method according to claim 1 further comprising accepting and associating ground truth input by the user with said datasets.

13. A method according to claim 12 wherein the database stores ground truth associated with the stored datasets or processed datasets.

14. A method according to claim 1 wherein the database stores user truth based on the interpretation of the dataset by the user.

15. A method according to claim 1 wherein said data to mediate the processing and/or interpretation of said dataset is data relating to similar datasets stored in the database.

16. A method according to claim 1 wherein said data to mediate the processing and/or interpretation of said dataset is statistical data relating to the information domain.

17. A method according to claim 1 wherein the mediation of the processing of the dataset comprises updating image analysis parameters in the processing software.

18. A method according to claim 1 wherein the information in the database is accessible by reference to the processed dataset.

19. A method according to claim 1 wherein the processed dataset comprises quantitative values derived from the dataset.

20. A computer readable storage medium carrying a computer program comprising program code means for carrying out the method of claim 1 when said program is run on a computer.

21. Data processing apparatus operative to accept input of data from a user acquired from instrumentation as a dataset of an information domain which dataset comprises image data which comprises medical image data, to process the dataset under the control of processing software said processing comprising the input of “

user truth”

based on interpretation of the dataset by the user, to provide a processed dataset to assist in the further interpretation of the dataset, and operative to store via a network said dataset and said processed dataset into a database of information relating to said domain acquired from a plurality of users;

said software comprising a software agent operative, on the basis of at least one of;

attributes of said dataset, attributes of the processed dataset and user-input, to access the database to retrieve data or software to mediate the processing or interpretation of said dataset wherein the software agent is operative to transfer to the database data concerning manual adjustments made to said processed dataset by the user, and constituting said “

user truth”

, and wherein the system is operative to mediate the processing of said dataset by updating said processing software using at least one of data and software from the database.