Patient data mining

• US 9,165,116 B2
• Filed: 12/15/2014
• Issued: 10/20/2015
• Est. Priority Date: 11/02/2001
• Status: Expired due to Term

• Current Assignee: Cerner Innovation, Inc. (Oracle Corporation)
• Original Assignee: Cerner Innovation, Inc. (Oracle Corporation)
• Inventors: Amies, Christopher Jude, Goel, Arun Kumar, Niculescu, Radu Stefan, Sandilya, Sathyakama, Rao, Bharat R, Warrick, Thomas R
• Primary Examiner(s): ALHIJA, SAIF A
• Application Number:
  US14/569,870
• Publication Number:
  US 20150100352A1
• Time in Patent Office: 309 Days
• Field of Search: 703/2
• US Class Current: 1/1
• CPC Class Codes: G06F 16/30 of unstructured textual data document management systems G06F16/93
• CPC Class Codes: G06Q 10/10 Office automation; Time management
• CPC Class Codes: G16H 10/20 for electronic clinical trials or questionnaires
• CPC Class Codes: G16H 10/60 for patient-specific data, e.g. for electronic patient records
• CPC Class Codes: G16H 15/00 ICT specially adapted for medical reports, e.g. generation or transmission thereof
• CPC Class Codes: G16H 40/20 for the management or administration of healthcare resources or facilities, e.g. managing hospital staff or surgery rooms
• CPC Class Codes: G16H 40/63 for local operation
• CPC Class Codes: G16H 50/20 for computer-aided diagnosis, e.g. based on medical expert systems
• CPC Class Codes: G16H 50/30 for calculating health indices; for individual health risk assessment
• CPC Class Codes: G16H 50/50 for simulation or modelling of medical disorders
• CPC Class Codes: G16H 50/70 for mining of medical data, e.g. analysing previous cases of other patients
• CPC Class Codes: G16H 70/60 relating to pathologies
• CPC Class Codes: Y10S 128/92 Computer assisted medical diagnostics

Abstract

The present invention provides a data mining framework for mining high-quality structured clinical information. The data mining framework includes a data miner that mines medical information from a computerized patient record (CPR) based on domain-specific knowledge contained in a knowledge base. The data miner includes components for extracting information from the CPR, combining all available evidence in a principled fashion over time, and drawing inferences from this combination process. The mined medical information is stored in a structured CPR which can be a data warehouse.

Claims

1. A system for producing structured clinical information from patient records, the system comprising:

a memory configured to store a patient record having patient information including unstructured data and structured data;

a probabilistic data miner of a computer platform configured to (a) mine the structured and unstructured information for multiple pieces of data related to a variable for the patient, (b) combine the multiple pieces of data related to the variable into a value of the variable for the patient, the value being a function of the multiple pieces of data related to the variable, and (c) repeat (a) and (b) for a plurality of different variables of the same patient to find values for the variables representing the patient at a same time, the multiple pieces of data for one variable of the different variables being different than the multiple pieces of data for another one of the different variables;

wherein one or both of (a) and (b) are performed as a function of domain-specific criteria.

2. The system of claim 1 wherein the probabilistic data miner is configured to extract the multiple pieces of data as a function of the domain-specific criteria.

3. The system of claim 1 wherein the probabilistic data miner is configured to combine the multiple pieces of data for the one variable as a function of the domain-specific criteria, the domain-specific criteria comprising knowledge about a disease.

4. The system of claim 1 wherein the probabilistic data miner comprises an extraction component for extracting the multiple pieces of data and outputting a probabilistic assertion.

5. The system of claim 1 wherein the probabilistic data miner is configured to infer a patient state as a function of the values for the different variables.

6. The system of claim 5 wherein the probabilistic data miner is configured to infer the patient state as a function of probabilities, one of the probabilities assigned to each of the values.

7. The system of claim 5 wherein the inference is a function of a statistical model of a pattern of evolution of a disease across a patient population and the relationship between a patient'"'"'s disease and observed variables.

8. The system of claim 1 wherein the patient information includes one or more of:

medical information, financial information, demographic information or combinations thereof, the unstructured data including two or more of;

the free text information, medical image information, laboratory information, prescription drug information, waveform information or combinations thereof.

9. The system of claim 1 wherein the probabilistic data miner is configured to extract key phrases from the unstructured data, the unstructured data comprising free text treatment notes, the key phrases comprising at least part of the domain-specific criteria.

10. The system of claim 1 wherein the probabilistic data miner is configured to output structured clinical information with probability information from the multiple pieces of data from the unstructured data.

11. The system of claim 1 wherein the domain-specific criteria includes institution-specific domain knowledge, disease-specific domain knowledge, or combinations thereof.

12. The system of claim 1 wherein the unstructured data includes two or more of:

ASCII text strings, image information in DICOM format, text documents, or combinations thereof, partitioned based on domain knowledge.

13. The system of claim 1 wherein the probabilistic data miner is run at arbitrary intervals, periodic intervals, in an online mode, or combinations thereof.

14. The system of claim 1 wherein a repository interface is used to access at least some of the patient information used by the probabilistic data miner, wherein the repository interface is a configurable data interface, which varies depending on hospital.

15. A method for providing structured clinical information from patient records, the method comprising:

mining, by a processor, unstructured free text in a physician'"'"'s note of a patient record from a healthcare provider, the mining comprising extracting evidence of a first value of a first variable;

creating, probabilistically by the processor, structured clinical data of values for respective multiple variables including the first value of the first variable, the structured clinical data for each of the multiple values created from multiple sources of mined evidence, the evidence of the first value being one of the multiple sources of mined evidence for the first variable; and

providing the mining of the unstructured free text as a service to the healthcare provider;

outputting at least one of the multiple values from the structured clinical data as a function of an input question regarding a patient.

16. The method of claim 15 wherein mining comprises mining the patient record of the healthcare provider and a patient record of another healthcare provider, and wherein providing comprises providing the mining as the service to the healthcare provider and the other healthcare provider.

17. The method of claim 15 wherein providing comprises providing the service by a third party service provider.

18. The method of claim 15 wherein creating comprises:

combining a set of probabilistic assertions into one or more unified probabilistic assertions; and

inferring a patient state from the one or more unified probabilistic assertions; and

wherein outputting comprises communicating the patient state.

19. The method of claim 15 wherein mining comprises mining as a function of institution-specific domain knowledge of the healthcare provider.

20. The method of claim 15 wherein mining comprises mining structured and unstructured data sources for the mined evidence.