System and method for evaluating data sets over a communications network

• US 7,779,025 B2
• Filed: 12/12/2005
• Issued: 08/17/2010
• Est. Priority Date: 12/13/2004
• Status: Expired due to Fees

• Current Assignee: U.S. Army Medical Research And Materiel Command, United States Army (U.S. Department Of Defense)
• Original Assignee: The United States Of America As Represented By The Secretary Of The Army
• Inventors: Gunawardena, Jamila, Reifman, Jacques
• Primary Examiner(s): Pham; Khanh B
• Assistant Examiner(s): Ng; Amy
• Application Number:
  US11/298,606
• Publication Number:
  US 20060202037A1
• Time in Patent Office: 1,709 Days
• Field of Search: 707/10, 707/1, 707/3, 707/999.001, 707/999.003, 707/999.01, 707/766, 707/769, 600/300
• US Class Current: 707/766
• CPC Class Codes: G16H 15/00 ICT specially adapted for medical reports, e.g. generation or transmission thereof
• CPC Class Codes: G16H 40/67 for remote operation

Abstract

The present invention is directed to a system and method for providing researchers with a computer-assisted tool to evaluate, over a communications network, large electronic data sets stored on a remote server. The system allows the centralization of multiple data manipulation and analysis software modules co-located with the large data sets to reduce the amount of data from the data sets and software that has to be downloaded to a particular user'"'"'s client computer. The system is designed to accommodate multiple researchers located at different geographic locations each of whom may have, with regard to the data, different scientific research objectives.

Claims

1. A computer network including a computer-implemented data evaluation tool for evaluating electronic data sets and capable of displaying an output of the evaluation on a display medium, comprising:

a communications network;

a first server in data communication with the communications network;

a first electronic database stored on the first server and populated with data records in a plurality of record series, the data records including a plurality of attribute data records containing data fields populated with discrete information and at least a plurality of time-series of data records containing data fields populated with time-series information;

a plurality of analysis modules on the first server, each of the plurality of analysis modules adapted to input a set of client-customized operations to the record series in the database and output results that satisfy the operations, the analysis modules providing mathematical manipulation of attribute and/or time-series data to generate new time-series data or a scalar result, the mathematical manipulation comprising at least one mathematical manipulation that integrates two or more of the record series, the analysis modules comprising a data query module capable of querying across different studies in the database and retrieve specific data subsets that meet client-defined criteria, and the analysis modules comprising a data analysis module having an ability to perform mathematical manipulations of the attribute and/or time-series data on the data subsets satisfying the query results;

a data analysis customizing function, permitting the client to customize in an interactive, arbitrary manner the data query module and the data analysis module by selecting study data from the database and analysis function operators from a library of functions and arranging them in a desired, sequential order, wherein the analysis function operators perform mathematical manipulations on the study data for generating the result;

at least one of said analysis customization, utilizing the data query module, wherein the client customizes the query using at least one of the study data from the database and a plurality of analysis function operators from the library of functions sequentially arranging the data and the functions as a function chain, so that the output of one function serves as the input to a subsequent function, the function chain effecting a selection based on client-supplied constraints applied to attribute and/or time-series data, resulting in the retrieval of data subsets that satisfy client query customization intent;

at least one of said analysis customization functions utilizing the data analysis module, wherein the client customizes analysis using at least one of the study data from the database and a plurality of analysis function operators from the library of functions sequentially arranging the data and the functions as a function chain, so that the output of one function serves as the input to a subsequent function, the function chain operating on the data subsets retrieved by the query module and using client-supplied constraints to perform mathematical operations to attribute and/or time-series data, resulting in derivative scalar and/or time-series data that satisfy client analysis customization intent applied to results of query module; and

a first interface associated with a first client computer in data communication with the communications network, the first interface adapted to sending a request to one of the plurality of analysis modules and adapted to displaying on the first client computer the result.

2. The data evaluation tool according to claim 1, wherein the first interface has a capability of selectively displaying at least some of the time-series data records or said result after application of analytical procedures to the time-series data.

3. The data evaluation tool according to claim 2, wherein the time-series data records one of EKG, SpO2, respiratory rate, NIBP, end-tidal CO2, and heart rate time-series data.

4. The data evaluation tool according to claim 1, wherein the plurality of analysis modules comprises at least one of a data visualization module and data cleansing module.

5. The data evaluation tool according to claim 1, further comprising:

a second interface associated with a second client computer in data communication with the communications network, the second interface adapted to sending a request to one of the plurality of analysis modules and displaying on the second client computer the result, thereby eliminating a need to download tools and data from the first server to the client computers;

the data evaluation tool providing an architecture that facilitates on-line collaboration among different users, allows for inter-study comparisons; and

the data evaluation tool further managing the database records.

6. The data evaluation tool according to claim 1, further comprising a library of data analysis functions stored on the first server, wherein each of the analysis functions comprises an analysis function operator for generating the result.

7. The data evaluation tool according to claim 6, comprising providing the analysis function operator as a univariate or multivariate data operator.

8. The data evaluation tool according to claim 1, further comprising a second database stored on the first server and populated with data records, wherein the records in the first database differ from the records in the second database, thereby permitting multiple databases to generate the new time-series data or scalar result.

9. The data evaluation tool according to claim 1, further comprising the analysis module receiving plural data sets, at least one of said data sets includes data gathered at a different sampling frequency and having a different number of data points from another one of said data sets, thereby permitting multiple databases to generate the new time-series data or scalar result.

10. The data evaluation tool according to claim 9, further comprising the analysis modules obtaining further data related to the source of the data sets, said further data used to analyze the data sets.

11. The data evaluation tool according to claim 1, further comprising:

a library of data analysis functions stored on the first server, wherein each of the analysis functions comprises an analysis function operator for generating the result;

at least one of said analysis functions stored on the first server selected based on the content of the request providing an analysis function from the library of data analysis functions, and in the case of plural functions, sequentially executing the functions as a function chain, the function chain effecting a selection based on client supplied constraints applied to attribute and/or time-series data, of a subset of subjects from the database, and effecting a mathematical analysis of time-series data associated with each of the selected subjects.

12. The data evaluation tool according to claim 1, further comprising:

the data analysis customizing function comprising a library of data analysis functions stored on the first server, wherein each of the analysis functions comprises an analysis function operator for generating the result;

at least one of said analysis functions stored on the first server selected based on the content of the request providing an analysis function from the library of data analysis functions, and in the case of plural functions, sequentially executing the functions as a function chain, the function chain effecting a selection based on client supplied constraints applied to attribute and/or time-series data, of a subset of subjects from the database, and effecting a mathematical analysis of time-series data associated with each of the selected subjects, the analysis selectively comprising standard database queries on attribute data, and capable of selecting one of directly query based on the original time-series data, or results after application of analytical procedures to the time-series data.

13. The data evaluation tool according to claim 1, further comprising:

the data analysis customizing function comprising a library of data analysis functions stored on the first server, wherein each of the analysis functions comprises an analysis function operator for generating the result; and

at least one of the analysis modules able implement individual functions are selected in sequence from a library of functions in a function chain to permit the user to customize a sequence of data operations by selecting functions from the library of functions and arrange the functions in a desired order.

14. A communications network-enabled tool for evaluating electronic data sets in a distributed computing environment, comprising:

a server;

an electronic database in data communications with the server and populated with data records in a plurality of record series, the data records including a plurality of attribute data records containing data fields populated with discrete information and at least a plurality of time-series of data records containing data fields populated with time-series information;

a plurality of analysis modules associated with the server and adapted to input a set of client-customized operations to the record series in the database and output results that satisfy the operations, the analysis modules providing mathematical manipulation of attribute and/or time-series data to generate new time-series data or a scalar result, the mathematical manipulation comprising at least one mathematical manipulation that integrates two or more of the record series, the analysis modules comprising a data query module capable of querying across different studies in the database and retrieve specific data subsets that meet client-defined criteria, and the analysis modules comprising a data analysis module having an ability to perform mathematical manipulations of the attribute and/or time-series data on the data subsets satisfying the query results;

a data analysis customizing function, permitting the client to customize in an interactive, arbitrary manner the data query module and the data analysis module by selecting study data from the database and analysis function operators from a library of functions and arranging them in a desired, sequential order, wherein the analysis function operators perform mathematical manipulations on the study data for generating the result;

at least one of said analysis customization functions utilizing the data query module, wherein the client customizes the query using at least one of the study data from the database and a plurality of analysis function operators from the library of functions sequentially arranging the data and the functions as a function chain, so that the output of one function serves as the input to a subsequent function, the function chain effecting a selection based on client-supplied constraints applied to attribute and/or time-series data, resulting in the retrieval of data subsets that satisfy client query customization intent;

at least one of said analysis customization, utilizing the data analysis module, wherein the client customizes analysis using at least one of the study data from the database and a plurality of analysis function operators from the library of functions sequentially arranging the data and the functions as a function chain, so that the output of one function serves as the input to a subsequent function, the function chain operating on the data subsets retrieved by the query module and using client-supplied constraints to perform mathematical operations to attribute and/or time-series data, resulting in derivative scalar and/or time-series data that satisfy client analysis customization intent applied to results of query module;

a plurality of client computers in data communication with the server adapted to sending a request to one of the plurality of analysis modules; and

an interface associated with each of the plurality client computers for displaying the result on at least one of the plurality of client computers.

15. The evaluation tool according to claim 14, wherein:

the parameters of the predetermined class provide early indicators of a clinical outcome; and

the data records are time-series data records containing data fields populated with physiological time-series information.

16. The evaluation tool according to claim 14, wherein the plurality of analysis modules comprises at least one of a data visualization module and data cleansing module.

17. The evaluation tool according to claim 14, further comprising a library of data analysis functions stored on the server, wherein each of the analysis functions comprises an analysis function operator for generating the result.

18. The evaluation tool according to claim 14, further comprising the analysis module receiving plural data sets, at least one of said data sets includes data gathered at a different sampling frequency and having a different number of data points from another one of said data sets, thereby permitting multiple databases to generate the new time-series data or scalar result.

19. The evaluation tool according to claim 18, further comprising the analysis modules obtaining further data related to the source of the data sets, said further data used to analyze the data sets.

20. The evaluation tool according to claim 14, further comprising:

a library of data analysis functions stored on the first server, wherein each of the analysis functions comprises an analysis function operator for generating the result;

at least one of said analysis functions stored on the first server selected based on the content of the request providing an analysis function from the library of data analysis functions, and in the case of plural functions, sequentially executing the functions as a function chain, the function chain effecting a selection based on client supplied constraints applied to attribute and/or time-series data, of a subset of subjects from the database, and effecting a mathematical analysis of time-series data associated with each of the selected subjects.

21. The evaluation tool according to claim 14, further comprising:

the data analysis customizing function comprising a library of data analysis functions stored on the first server, wherein each of the analysis functions comprises an analysis function operator for generating the result;

at least one of said analysis functions stored on the first server selected based on the content of the request providing an analysis function from the library of data analysis functions, and in the case of plural functions, sequentially executing the functions as a function chain, the function chain effecting a selection based on client supplied constraints applied to attribute and/or time-series data, of a subset of subjects from the database, and effecting a mathematical analysis of time-series data associated with each of the selected subjects, the analysis selectively comprising standard database queries on attribute data, and capable of selecting one of directly query based on the original time-series data, or results after application of analytical procedures to the time-series data.

22. The evaluation tool according to claim 14, further comprising:

the data analysis customizing function comprising a library of data analysis functions stored on the first server, wherein each of the analysis functions comprises an analysis function operator for generating the result; and

at least one of the analysis modules able implement individual functions are selected in sequence from a library of functions in a function chain to permit the user to customize a sequence of data operations by selecting functions from the library of functions and arrange the functions in a desired order.

23. A method for evaluating electronic data records over an electronic communications network, comprising the steps of:

receiving over a communications network a set of electronic data records;

storing the set of electronic data records in a plurality of record series in at least one database on a server, the data records including a plurality of attribute data records containing data fields populated with discrete information and at least a plurality of time-series of data records containing data fields populated with time-series information;

receiving at the server a first instruction from a remote client computer to identify one or more of the data records, the data records including time-series data records containing data fields populated with time-series information;

receiving at the server a second instruction to input a set of client-customized operations to be applied to the identified data records;

receiving at the server a third instruction to analyze the result of the applied mathematical operation, the analysis providing mathematical manipulation of attribute and/or time-series data to generate new time-series data or a scalar result;

formatting a response using at least the result from the applied mathematical operation and output results that satisfy the operations, the parameters providing early indicators of an evaluation outcome, the analysis allowing query across different studies in the database and retrieve specific data subsets that meet client-defined criteria, and the analysis able to perform mathematical manipulations of the attribute and/or time-series data on the data subsets satisfying the query results;

permitting the client to customize in an interactive, arbitrary manner a data query module and a data analysis module by selecting study data from the database and analysis function operators from a library of functions and arranging them in a desired, sequential order, wherein the analysis function operators perform mathematical manipulations on the study data for generating the result;

customizing the query using at least one of the study data from the database and a plurality of analysis function operators from the library of functions sequentially arranging the data and the functions as a function chain, so that the output of one function serves as the input to a subsequent function, the function chain effecting a selection based on client-supplied constraints applied to attribute and/or time-series data, resulting in the retrieval of data subsets that satisfy client query customization intent;

customizing analysis using at least one of the study data from the database and a plurality of analysis function operators from the library of functions sequentially arranging the data and the functions as a function chain, so that the output of one function serves as the input to a subsequent function, the function chain operating on the data subsets retrieved by the query and using client-supplied constraints to perform mathematical operations to attribute and/or time-series data, resulting in derivative scalar and/or time-series data that satisfy client analysis customization intent applied to results of query module; and

delivering or providing the result to the client computer, wherein the first, second, and third instructions are part of the same request or a different request.

24. The method according to claim 23, further comprising selectively displaying at least a portion of a time-series data set or the result after application of analytical procedures to the time-series data.

25. The method according to claim 24, wherein:

the parameters of the predetermined class provide early indicators of a clinical outcome; and

the time-series data set is a physiologic time-series data set.

26. The method according to claim 23, wherein the step of receiving a set of electronic data records comprises collecting electronic signals from a monitoring device.

27. The method according to claim 23, wherein the steps of receiving the instructions and delivering or providing the result are done over a secure HTTP connection between the server and the client computer, thereby facilitating on-line collaboration among different users and eliminating a need to download tools and data from the first server to the client computers.

28. The method according to claim 23, further comprising the step of:

receiving authentication information from a user; and

comparing the authentication information to credentials stored in the at least one database.

29. The method according to claim 23, further comprising the step of:

receiving a plurality of instructions from the remote client computer; and

outputting a customized analysis report.

30. A method for mining and analyzing electronic data records over a distributed network of client computers, comprising the steps of:

storing on a central server a set of data records obtained from an electronic monitoring device in a plurality of record series;

receiving a data mining request, a function request, and an analysis request from a remote client computer;

selecting a subset of the data records based on the data mining request, the data records including a plurality of attribute data records containing data fields populated with discrete information and at least one a plurality of time-series data records containing data fields populated with time-series information;

performing at least one mathematical operation on the subset of data records based on the function request;

performing an analysis of a result of the at least one mathematical operation based on the analysis request by identifying parameters of a predetermined class, the analysis providing mathematical manipulation of scalar or time-series data to generate new time-series data or a scalar result, the parameters providing early indicators of an evaluation outcome, the mathematical manipulation comprising at least one mathematical manipulation that integrates two or more of the record series, the analysis allowing query across different studies in a database and retrieve specific data subsets that meet client-defined criteria, and the analysis able to perform mathematical manipulations of the attribute and/or time-series data on the data subsets satisfying the query results;

a data analysis customizing function, permitting the client to customize in an interactive, arbitrary manner a data query module and a data analysis module by selecting study data from the database and analysis function operators from a library of functions and arranging them in a desired, sequential order, wherein the analysis function operators perform mathematical manipulations on the study data for generating the result;

customizing the query using at least one of the study data from the database and a plurality of analysis function operators from the library of functions sequentially arranging the data and the functions as a function chain, so that the output of one function serves as the input to a subsequent function, the function chain effecting a selection based on client-supplied constraints applied to attribute and/or time-series data, resulting in the retrieval of data subsets that satisfy client query customization intent;

customizing analysis using at least one of the study data from the database and a plurality of analysis function operators from the library of functions sequentially arranging the data and the functions as a function chain, so that the output of one function serves as the input to a subsequent function, the function chain operating on the data subsets retrieved by the query and using client-supplied constraints to perform mathematical operations to attribute and/or time-series data, resulting in derivative scalar and/or time-series data that satisfy client analysis customization intent applied to results of query module; and

delivering or providing a response to the client computer containing information about the subset of data records, the result of the at least one mathematical operation, and the analysis of the result.

31. The method for mining and analyzing electronic data records according to claim 30, wherein:

the parameters of the predetermined class provide early indicators of a clinical outcome; and

the subset of data records includes physiologic time-series data.

32. The method for mining and analyzing electronic data records according to claim 30, wherein the steps of receiving, selecting and performing are done by the central server.

33. The method for mining and analyzing electronic data records according to claim 30, further comprising the receiving plural data sets, at least one of said data sets includes data gathered at a different sampling frequency and having a different number of data points from another one of said data sets, thereby permitting multiple databases to generate the new time-series data or scalar result.

34. The method for mining and analyzing electronic data records according to claim 33, further comprising obtaining further data related to the source of the data sets to analyze the data sets.

35. The method for mining and analyzing electronic data records according to claim 30, further comprising:

storing a library of data analysis functions stored on the central server, wherein each of the analysis functions comprises an analysis function operator for generating the result;

providing an analysis based on the content of the request from the library of data analysis functions, and in the case of plural functions, sequentially executing the functions as a function chain, the function chain effecting a selection based on client supplied constraints applied to attribute and/or time-series data, of a subset of subjects from the database, and effecting a mathematical analysis of time-series data associated with each of the selected subjects.

36. The method for mining and analyzing electronic data records according to claim 30, further comprising:

the execution of the data analysis customizing function comprising storing a library of data analysis functions stored on the central server, wherein each of the analysis functions comprises an analysis function operator for generating the result;

providing an analysis based on the content of the request from the library of data analysis functions, and in the case of plural functions, sequentially executing the functions as a function chain, the function chain effecting a selection based on client supplied constraints applied to attribute and/or time-series data, of a subset of subjects from the database, and effecting a mathematical analysis of time-series data associated with each of the selected subjects, the analysis selectively comprising standard database queries on attribute data, and capable of selecting one of directly query based on the original time-series data, or results after application of analytical procedures to the time-series data.

37. The method for mining and analyzing electronic data records according to claim 30, further comprising:

the execution of the data analysis customizing function comprising storing data analysis functions as a library, wherein each of the analysis functions comprises an analysis function operator for generating the result; and

at least one of the analysis modules able implement individual functions are selected in sequence from a library of functions in a function chain to permit the user to customize a sequence of data operations by selecting functions from the library of functions and arrange the functions in a desired order.