Discovery routing systems and engines

• US 10,114,925 B2
• Filed: 07/28/2014
• Issued: 10/30/2018
• Est. Priority Date: 07/26/2013
• Status: Active Grant

• Current Assignee: Nant Holdings IP, LLC (NantWorks, LLC)
• Original Assignee: Nant Holdings IP, LLC (NantWorks, LLC)
• Inventors: Soon-Shiong, Patrick
• Primary Examiner(s): Holcomb, Mark
• Application Number:
  US14/445,025
• Publication Number:
  US 20150032468A1
• Time in Patent Office: 1,555 Days
• Field of Search: 705 2- 4
• CPC Class Codes: G16B 50/00 ICT programming tools or database systems specially adapted for bioinformatics
• CPC Class Codes: G16B 50/30 Data warehousing; Computing architectures
• CPC Class Codes: G16H 50/20 for computer-aided diagnosis, e.g. based on medical expert systems
• CPC Class Codes: G16H 80/00 ICT specially adapted for facilitating communication between medical practitioners or patients, e.g. for collaborative diagnosis, therapy or health monitoring
• CPC Class Codes: G16Z 99/00 Subject matter not provided for in other main groups of this subclass

Abstract

The inventive subject matter provides apparatus, systems, and methods that improve on the pace of discovering new practical information based on large amounts of datasets collected. In most cases, anomalies from the datasets are automatically identified, flagged, and validated by a cross-validation engine. Only validated anomalies are then associated with a subject matter expert who is qualified to take action on the anomaly. In other words, the inventive subject matter bridges the gap between the overwhelming amount of scientific data which can now be harvested and the comparatively limited amount analytical resources available to extract practical information from the data. Practical information can be in the form of trends, patterns, maps, hypotheses, or predictions, for example, and such practical information has implications in medicine, in environmental sciences, entertainment, travel, shopping, social interactions, or other areas.

Claims

1. A discovery routing system comprising:

a knowledge database programmed to store a plurality of datasets, each dataset comprising at least one descriptor-value pair having a descriptor and an associated value;

an analytical engine coupled with the knowledge database, and programmed to identify at least one anomaly in a dataset of the plurality of datasets, wherein the at least one anomaly is associated with a condition and characterizes a descriptor-value pair having a value that deviates from a qualifier associated with the descriptor beyond a predetermined threshold value for the descriptor;

a cross-validation engine comprising an association application and a relationship application, coupled with the analytical engine and programmed to designate the at least one anomaly as a significant anomaly by (i) in the association application, cross-referencing the at least one anomaly with one or more other descriptor-value pairs of attributes that are associated with the condition and known to have a value that deviates from the qualifier, wherein the cross-referencing includes traversing the dataset of the plurality of datasets to identify at least one of the one or more other descriptor-value pairs of attributes, (ii) identifying further attributes of the at least one anomaly by performing a comparison of at least one attribute with the at least one value of the descriptor-value pair of the at least one anomaly using at least one parameter different from a parameter of the qualifier, and (iii) in the relationship application, omitting the descriptor-value pair of the anomaly from a significant anomaly dataset based on the cross-referencing with the one or more other descriptor-value pairs of the further attributes when there is a priori knowledge showing a causal relationship or correlation between the further attributes, the at least one anomaly and the condition, and including the descriptor-value pair of the at least one anomaly in the significant anomaly dataset if at least one other descriptor-value pair is identified as related to the condition; and

a subscription engine comprising a subscriber database, coupled with the cross-validation engine and programmed to (i) subscribe a plurality of expert subscribers, wherein each expert subscriber is associated with an identifier in the subscriber database, and (ii) associate the descriptor-value pair in the significant anomaly dataset with an expert subscriber based on the identifier.

2. The discovery routing system of claim 1, wherein the cross-validation engine is further programmed to designate the at least one anomaly as a significant anomaly by (i) identifying the condition based on a priori knowledge, (ii) implying a relationship between the at least one anomaly and the condition by cross-referencing additional data in the dataset having a secondary association with the condition, and (iii) then designating the at least one anomaly as a significant anomaly.

3. The discovery routing system of claim 2, wherein the cross-validation engine is further programmed to designate the at least one anomaly as an insignificant anomaly when the a priori knowledge indicates a known relationship between the condition and the descriptor.

4. The discovery routing system of claim 2, wherein the relationship comprises a correlation relationship.

5. The discovery routing system of claim 2, wherein the relationship comprises a causal relationship.

6. The discovery routing system of claim 2, wherein the relationship comprises a pre-requisite relationship.

7. The discovery routing system of claim 1, wherein the dataset of the plurality of datasets includes data derived from at least one genetically related source.

8. The discovery routing system of claim 1, wherein the dataset of the plurality of datasets includes data derived from a single person.

9. The discovery routing system of claim 1, wherein the expert subscriber is at least one of the following:

a computer, an algorithm, an individual person, a group of people, a business entity, a government body, a publisher, and a trade association.

10. The discovery routing system of claim 1, wherein the predetermined threshold value for the descriptor is chosen according to at least one of the following:

an a priori standard, a statistically determined standard, a standard derived by an algorithm, a comparison with historical values, a comparison with boundary conditions, a predicted value, an analysis of confidence factors, and a user-defined standard.

11. The discovery routing system of claim 1, wherein the cross-validation engine is further programmed to receive a confirmation of the relationship between the condition and the at least one anomaly from the expert subscriber.

12. The discovery routing system of claim 1, wherein the subscription engine is further programmed to generate an association notification when the significant anomaly is associated with an expert subscriber.

13. The discovery routing system of claim 1, wherein the subscription engine is further programmed to transmit an assignment notification to the associated expert subscriber.

14. A computer related product comprising a non-transitory computer readable medium storing instructions that cause a processor to execute the steps of:

providing access to a knowledge database programmed to store a plurality of datasets, each dataset comprising at least one descriptor-value pair having a descriptor associated with a value;

providing a discovery routing engine that is coupled with the knowledge database;

associating, by the discovery routing engine, a qualifier with at least one descriptor;

identifying, by the discovery routing engine, at least one anomaly in a dataset of the plurality of datasets, wherein the at least one anomaly is associated with a condition and characterizes a descriptor-value pair having a value that deviates from a qualifier associated with the descriptor beyond a predetermined threshold value for the descriptor;

designating, by the discovery routing engine, the at least one anomaly as a significant anomaly by (i) cross-referencing the at least one anomaly with one or more other descriptor-value pairs of attributes that are associated with the condition and known to have a value that deviates from the qualifier, wherein the cross-referencing includes traversing the dataset of the plurality of datasets to identify at least one of the one or more other descriptor-value pairs of attributes, (ii) identifying further attributes of the at least one anomaly by performing a comparison of at least one attribute with the at least one value of the descriptor-value pair of the at least one anomaly using at least one parameter different from a parameter of the qualifier, and (iii) omitting the descriptor-value pair of the anomaly from a significant anomaly dataset based on the cross-referencing with the one or more other descriptor-value pairs of the further attributes when there is a priori knowledge showing a causal relationship or correlation between the further attributes, the at least one anomaly and the condition and including the descriptor-value pair of the at least one anomaly in the significant anomaly dataset if at least one other descriptor-value pair is identified as related to the condition;

subscribing, by the discovery routing engine, a plurality of expert subscribers, wherein each expert subscriber is associated with an identifier in a subscriber database; and

associating, by the discovery routing engine, the descriptor-value pair in the significant anomaly dataset with an expert subscriber based on the identifier.

15. The computer related product of claim 14, wherein the step of designating the at least one anomaly as a significant anomaly comprises:

identifying the condition based on a priori knowledge;

implying a relationship between the at least one anomaly and the condition by cross-referencing additional data in the dataset having a secondary association with the condition; and

then designating the at least one anomaly as a significant anomaly.

16. The computer related product of claim 15, wherein the step of designating the at least one anomaly as a significant anomaly comprises designating the at least one anomaly as an insignificant anomaly when the a priori knowledge indicates a known relationship between the condition and the descriptor.

17. The computer related product of claim 15, wherein the relationship comprises a correlation relationship.

18. The computer related product of claim 15, wherein the relationship comprises a causal relationship.

19. The computer related product of claim 15, wherein the relationship comprises a pre-requisite relationship.

20. The computer related product of claim 14, wherein the dataset of the plurality of datasets includes data derived from at least one genetically related source.

21. The computer related product of claim 14, wherein the dataset of the plurality of datasets includes data derived from a single person.

22. The computer related product of claim 14, wherein the expert subscriber is at least one of the following:

a computer, an algorithm, an individual person, a group of people, a business entity, a government body, a publisher, and a trade association.

23. The computer related product of claim 14, wherein the predetermined threshold value for the descriptor is chosen according to at least one of the following:

an a priori standard, a statistically determined standard, a standard derived by an algorithm, a comparison with historical values, a comparison with boundary conditions, a predicted value, an analysis of confidence factors, and a user-defined standard.

24. The computer related product of claim 14, wherein the instructions further cause the processor to execute the step of receiving a confirmation of the relationship between the condition and the at least one anomaly from the expert subscriber.

25. The computer related product of claim 14, wherein instructions further cause the processor to execute the step of generating an association notification when the significant anomaly is associated with an expert subscriber.

26. The computer related product of claim 14, wherein the instructions further cause the processor to execute the step of transmitting an assignment notification to the associated expert subscriber.

27. A method of routing discoveries, the method comprising:

providing access to a knowledge database programmed to store a plurality of datasets, each dataset comprising at least one descriptor-value pair having a descriptor associated with a value;

providing a discovery routing engine that is coupled with the knowledge database;

associating, by the discovery routing engine, a qualifier with at least one descriptor;

identifying, by the discovery routing engine, at least one anomaly in a dataset of the plurality of datasets, wherein the at least one anomaly is associated with a condition and characterizes a descriptor-value pair having a value that deviates from a qualifier associated with the descriptor beyond a predetermined threshold value for the descriptor;

designating, by the discovery routing engine, the at least one anomaly as a significant anomaly by (i) cross-referencing the at least one anomaly with one or more other descriptor-value pairs of attributes that are associated with the condition and known to have a value that deviates from the qualifier, wherein the cross-referencing includes traversing the dataset of the plurality of datasets to identify at least one of the one or more other descriptor-value pairs of attributes, (ii) identifying further attributes of the at least one anomaly by performing a comparison of at least one attribute with the at least one value of the descriptor-value pair of the at least one anomaly using at least one parameter different from a parameter of the qualifier, and (iii) omitting the descriptor-value pair of the anomaly from a significant anomaly dataset based on the cross-referencing with the one or more other descriptor-value pairs of the further attributes when there is a priori knowledge showing a causal relationship or correlation between the further attributes, the at least one anomaly and the condition and including the descriptor-value pair of the at least one anomaly in the significant anomaly dataset if at least one other descriptor-value pair is identified as related to the condition;

subscribing, by the discovery routing engine, a plurality of expert subscribers, wherein each expert subscriber is associated with an identifier in a subscriber database; and

associating, by the discovery routing engine, the descriptor-value pair in the significant anomaly dataset with an expert subscriber based on the identifier.

28. The method of claim 27, wherein designating the at least one anomaly as a significant anomaly by cross-referencing additional data in the dataset comprising:

identifying the condition based on a priori knowledge;

implying a relationship between the at least one anomaly and the condition by cross-referencing additional data in the dataset having a secondary association with the condition; and

then designating the at least one anomaly as a significant anomaly.

29. The method of claim 28, wherein designating the at least one anomaly as an insignificant anomaly comprises designating the at least one anomaly as an insignificant anomaly when the a priori knowledge indicates a known relationship between the condition and the descriptor.

30. The method of claim 28, wherein the relationship comprises a correlation relationship.

31. The method of claim 28, wherein the relationship comprises a causal relationship.

32. The method of claim 28, wherein the relationship comprises a pre-requisite relationship.

33. The method of claim 27, wherein the dataset of the plurality of datasets includes data derived from at least one genetically related source.

34. The method of claim 27, wherein the dataset of the plurality of datasets includes data derived from a single person.

35. The method of claim 27, wherein the expert subscriber is at least one of the following:

a computer, an algorithm, an individual person, a group of people, a business entity, a government body, a publisher, and a trade association.

36. The method of claim 27, wherein the predetermined threshold value for the descriptor is chosen according to at least one of the following:

an a priori standard, a statistically determined standard, a standard derived by an algorithm, a comparison with historical values, a comparison with boundary conditions, a predicted value, an analysis of confidence factors, and a user-defined standard.

37. The method of claim 27, further comprising receiving a confirmation of the relationship between the condition and the at least one anomaly from the expert subscriber.

38. The method of claim 27, further comprising generating an association notification when the significant anomaly is associated with an expert subscriber.

39. The method of claim 27, further comprising transmitting an assignment notification to the associated expert subscriber.