Automated learning for medical data processing system

• US 8,612,261 B1
• Filed: 12/17/2012
• Issued: 12/17/2013
• Est. Priority Date: 05/21/2012
• Status: Expired due to Fees

• Current Assignee: Health Management Associates Incorporated
• Original Assignee: Health Management Associates Incorporated
• Inventors: Chatfield, Kenneth R., Smith, Christopher B., Swanson, Brian S.
• Primary Examiner(s): PAULS, JOHN A
• Application Number:
  US13/717,469
• Time in Patent Office: 365 Days
• Field of Search: 705/3, 705/34, 704/9
• US Class Current: 705/3
• CPC Class Codes: G16H 10/60 for patient-specific data, e.g. for electronic patient records
• CPC Class Codes: G16H 40/67 for remote operation
• CPC Class Codes: G16H 50/20 for computer-aided diagnosis, e.g. based on medical expert systems
• CPC Class Codes: Y02A 90/10 Information and communication technologies [ICT] supporting adaptation to climate change, e.g. for weather forecasting or climate simulation

Abstract

In at least one embodiment, an automated medical data machine learning system and method allow a natural language processing (“NLP”) system to automatically learn via, for example, feedback to improve ongoing performance of the natural language processing system. The particular technology for improving the interpretation by the NLP system of future input data is a matter of design choice. In at least one embodiment, the automated medical data learning system and method includes a linguistics module that determines the part of speech of a particular term or term, such as use as a noun or verb. In at least one embodiment, the system and method learns an interpretation based on a source of the input data. In at least one embodiment, the system and method includes a statistics module that allows the system and method to determine a most probable interpretation or multiple interpretations.

Claims

1. An method comprising:

receiving clinician note data in a first electronic system that includes patient medical-related data;

processing the clinician note data using a natural language data processor and medical record interpretation rules to automatically convert the clinician note data into an electronic medical record;

sending the electronic medical record to a second electronic system to allow review and modification of the converted clinician note data in the electronic medical record;

receiving feedback data in the first electronic system that indicates modifications to the converted clinician note data in the electronic medical record;

processing the feedback data to identify modifications to the converted clinician note data in the electronic medical record;

analyzing the modifications in the feedback data to determine whether to modify the medical record interpretation rules; and

modifying the medical record interpretation rules in accordance with at least a portion of the feedback data for use in processing subsequently received clinician note data if analysis of the modifications in the feedback data determines to modify the medical record interpretation rules.

2. The method of claim 1 further comprising:

using the natural language processor to process the clinician note data in accordance with the modified medical record interpretation rules.

3. The method of claim 1 wherein:

receiving feedback data comprises receiving feedback data from one or more data sources in a group consisting of;

medical core measures feedback data and clinician feedback data; and

processing the feedback data to identify modifications to the converted clinician note data in the electronic medical record comprises at least one member of a group consisting of;

identifying modifications to the converted clinician note data in the electronic medical record as modifications from a medical core measures electronic system; and

identifying modifications to the converted clinician note data in the electronic medical record as modifications from a clinician medical note data entry and display system.

4. The method of claim 1 wherein:

receiving feedback data comprises receiving feedback data from multiple data sources in a group consisting of;

medical coding feedback data, medical core measures feedback data, and clinician feedback data; and

processing the feedback data to identify modifications to the converted clinician note data in the electronic medical record comprises at least one member of a group consisting of;

identifying modifications to the converted clinician note data in the electronic medical record as modifications from a medical coding electronic system;

identifying modifications to the converted clinician note data in the electronic medical record as modifications from a medical core measures electronic system; and

identifying modifications to the converted clinician note data in the electronic medical record as modifications from a clinician medical note data entry and display system.

5. The method of claim 1 wherein processing the feedback data to identify modifications to the converted clinician note data in the electronic medical record comprises:

analyzing linguistics of the electronic medical record as modified to determine discrete contexts of portions of the electronic medical record;

analyzing statistics associated with the medical record interpretation rules; and

modifying the medical record interpretation rules in accordance with the linguistics and statistics analysis.

6. The method of claim 5 wherein analyzing statistics associated with the medical record interpretation rules comprises:

identifying variables that affect one or more of the medical record interpretation rules; and

modifying the medical record interpretation rules so that processing of subsequent clinician note data has a higher probability of accuracy and completeness relative to a probability of accuracy and completeness in processing of previous clinician note data.

7. The method of claim 1 wherein the medical record interpretation rules are customized in accordance to a member of a group consisting of:

an individual clinician, a clinician'"'"'s medical specialty, an individual patient, a patient'"'"'s demographics, a medical facility, and a medical facility location.

8. The method of claim 1 further comprising:

identifying a medical record template corresponding to the clinician note data;

processing the clinician note data to standardize the clinician note data per predetermined data standards; and

operating the natural language processor to insert the clinician note data into the medical record template.

9. The method of claim 1 wherein the medical record interpretation rules comprise analytics associated with at least one member of the group of:

word count;

parts of speech;

medical ontologies; and

metadata associated with at least one of;

a clinician and a medical facility.

10. An apparatus comprising:

one or more processors; and

a memory, coupled to the one or more processors, and storing code therein that is executable by the one or more processors for;

receiving clinician note data in a first electronic system that includes patient medical-related data;

processing the clinician note data using a natural language data processor and medical record interpretation rules to automatically convert the clinician note data into an electronic medical record;

sending the electronic medical record to a second electronic system to allow review and modification of the converted clinician note data in the electronic medical record;

receiving feedback data in the first electronic system that indicates modifications to the converted clinician note data in the electronic medical record;

processing the feedback data to identify modifications to the converted clinician note data in the electronic medical record;

analyzing the modifications in the feedback data to determine whether to modify the medical record interpretation rules; and

modifying the medical record interpretation rules in accordance with at least a portion of the feedback data for use in processing subsequently received clinician note data if analysis of the modifications in the feedback data determines to modify the medical record interpretation rules.

11. The apparatus of claim 10 wherein the code is further executable by the one or more processors for:

using the natural language processor to process the clinician note data in accordance with the modified medical record interpretation rules.

12. The apparatus of claim 10 wherein:

to receive feedback data, the code is further executable by the one or more processors for receiving feedback data from one or more data sources in a group consisting of;

medical core measures feedback data and clinician feedback data; and

processing the feedback data to identify modifications to the converted clinician note data in the electronic medical record comprises at least one member of a group consisting of;

identifying modifications to the converted clinician note data in the electronic medical record as modifications from a medical core measures electronic system; and

identifying modifications to the converted clinician note data in the electronic medical record as modifications from a clinician medical note data entry and display system.

13. The apparatus of claim 10 wherein:

to receive feedback data, the code is further executable by the one or more processors for receiving feedback data from multiple data sources in a group consisting of;

medical coding feedback data, medical core measures feedback data, and clinician feedback data; and

processing the feedback data to identify modifications to the converted clinician note data in the electronic medical record comprises at least one member of a group consisting of;

identifying modifications to the converted clinician note data in the electronic medical record as modifications from a medical coding electronic system;

identifying modifications to the converted clinician note data in the electronic medical record as modifications from a medical core measures electronic system; and

identifying modifications to the converted clinician note data in the electronic medical record as modifications from a clinician medical note data entry and display system.

14. The apparatus of claim 10 wherein to process the feedback data to identify modifications to the converted clinician note data in the electronic medical record, the code is further executable by the one or more processors for:

analyzing linguistics of the converted clinician note data in the electronic medical record as modified to determine discrete contexts of portions of the electronic medical record;

analyzing statistics associated with the medical record interpretation rules; and

modifying the medical record interpretation rules in accordance with the linguistics and statistics analysis.

15. The apparatus of claim 14 wherein to analyze statistics associated with the electronic medical record interpretation rules, the code is further executable by the one or more processors for:

identifying variables that affect one or more of the medical record interpretation rules; and

modifying the medical record interpretation rules so that processing of subsequent clinician note data has a higher probability of accuracy and completeness relative to a probability of accuracy and completeness in processing of previous clinician note data.

16. The apparatus of claim 10 wherein the medical record interpretation rules are customized in accordance to a member of a group consisting of:

an individual clinician, a clinician'"'"'s medical specialty, an individual patient, a patient'"'"'s demographics, a medical facility, and a medical facility location.

17. The apparatus of claim 10 wherein the code is further executable by the one or more processors for:

identifying a medical record template corresponding to the clinician note data;

processing the clinician note data to standardize the clinician note data per predetermined data standards; and

operating the natural language processor to insert the clinician note data into the medical record template.

18. The apparatus of claim 10 wherein the medical record interpretation rules comprise analytics associated with at least one member of the group of:

word count;

parts of speech;

medical ontologies; and

metadata associated with at least one of;

a clinician and a medical facility.

19. A tangible, non-transitory computer readable medium comprising code stored therein and executable by one or more processors for:

receiving clinician note data in a first electronic system that includes patient medical-related data;

processing the clinician note data using a natural language data processor and medical record interpretation rules to automatically convert the clinician note data into an electronic medical record;

sending the electronic medical record to a second electronic system to allow review and modification of the converted clinician note data in the electronic medical record;

receiving feedback data in the first electronic system that indicates modifications to the converted clinician note data in the electronic medical record;

processing the feedback data to identify modifications to the converted clinician note data in the electronic medical record;

analyzing the modifications in the feedback data to determine whether to modify the medical record interpretation rules; and

modifying the medical record interpretation rules in accordance with at least a portion of the feedback data for use in processing subsequently received clinician note data if analysis of the modifications in the feedback data determines to modify the medical record interpretation rules.

20. The tangible, non-transitory computer readable medium of claim 19 wherein the code is further executable by the one or more processors for:

using the natural language processor to process the clinician note data in accordance with the modified medical record interpretation rules.

21. The tangible, non-transitory computer readable medium of claim 19 wherein:

to receive feedback data, the code is further executable by the one or more processors for receiving feedback data from one or more data sources in a group consisting of;

medical core measures feedback data and clinician feedback data; and

processing the feedback data to identify modifications to the converted clinician note data in the electronic medical record comprises at least one member of a group consisting of;

identifying modifications to the converted clinician note data in the electronic medical record as modifications from a medical core measures electronic system; and

identifying modifications to the converted clinician note data in the electronic medical record as modifications from a clinician medical note data entry and display system.

22. The tangible, non-transitory computer readable medium of claim 19 wherein:

to receive feedback data, the code is further executable by the one or more processors for receiving feedback data from multiple data sources in a group consisting of;

medical coding feedback data, medical core measures feedback data, and clinician feedback data; and

processing the feedback data to identify modifications to the converted clinician note data in the electronic medical record comprises at least one member of a group consisting of;

identifying modifications to the converted clinician note data in the electronic medical record as modifications from a medical coding electronic system;

identifying modifications to the converted clinician note data in the electronic medical record as modifications from a medical core measures electronic system; and

identifying modifications to the converted clinician note data in the electronic medical record as modifications from a clinician medical note data entry and display system.

23. The tangible, non-transitory computer readable medium of claim 19 wherein to process the feedback data to identify modifications to the converted clinician note data in the electronic medical record, the code is further executable by the one or more processors for:

analyzing linguistics of the converted clinician note data in the electronic medical record as modified to determine discrete contexts of portions of the electronic medical record;

analyzing statistics associated with the medical record interpretation rules; and

modifying the medical record interpretation rules in accordance with the linguistics and statistics analysis.

24. The tangible, non-transitory computer readable medium of claim 23 wherein to analyze statistics associated with the medical record interpretation rules, the code is further executable by the one or more processors for:

identifying variables that affect one or more of the medical record interpretation rules; and

modifying the medical record interpretation rules so that processing of subsequent clinician note data has a higher probability of accuracy and completeness relative to a probability of accuracy and completeness in processing of previous clinician note data.

25. The tangible, non-transitory computer readable medium of claim 19 wherein the medical record interpretation rules are customized in accordance to a member of a group consisting of:

an individual clinician, a clinician'"'"'s medical specialty, an individual patient, a patient'"'"'s demographics, a medical facility, and a medical facility location.

26. The tangible, non-transitory computer readable medium of claim 19 wherein the code is further executable by the one or more processors for:

identifying a medical record template corresponding to the clinician note data;

processing the clinician note data to standardize the clinician note data per predetermined data standards; and

operating the natural language processor to insert the clinician note data into the medical record template.

27. The tangible, non-transitory computer readable medium of claim 19 wherein the medical record interpretation rules comprise analytics associated with at least one member of the group of:

word count;

parts of speech;

medical ontologies; and

metadata associated with at least one of;

a clinician and a medical facility.