Automated clinical indicator recognition with natural language processing

Abstract

Computer-based, natural language processing systems and methods are provided for review of clinical documentation and other medical records, and for clinical documentation improvement. The systems and methods are configured to review documents in the record using a natural language processor and to identify clinical indicators with associated contextual information. The clinical indicators are compared to scenarios to generate markers based on an information model. The markers used to generate physician queries and other informational requests with supporting evidence for each query based on indicators identified in the record. In additional examples, pragmatic guidelines including business-based rules can also be utilized, either in combination with, or as part of, the scenarios in the information model.

1 . A computer-based, natural language processing system comprising:
 a computer processor;   a database in communication with the computer processor, the database comprising memory for storing a plurality of electronic documents relating to patient care of a patient; and   a natural language processor module executing on the computer processor, the natural language processor module configured to:
 review the plurality of electronic documents to identify clinical indicators and associated contextual information by synthesizing descriptions of historical data using multiple segments of text within the plurality of electronic documents, wherein the clinical indicators and associated contextual information are analyzed according to a document type in which the clinical indicators are identified and according to the associated contextual factors related to the clinical indicator; 
 compare the clinical indicators and associated contextual information to rule-based scenarios; 
 generate a marker based on the clinical indicators and associated contextual information, the marker identifying a record element associated with a medical condition relating to the patient care; 
 generate a query for a service provider, the query describing the marker; and 
 transmit the query to the service provider. 
   
     
     
         2 . The system of  claim 1 , wherein the record element indicates a medical condition of the patient absent from the plurality of electronic documents. 
     
     
         3 . The system of  claim 1 , wherein the marker identifies a medical condition, diagnosis, diagnosis code, or indicator of a medical condition or diagnosis that is associated with the patient, and which is absent from the plurality of electronic documents. 
     
     
         4 . The system of  claim 3 , wherein the marker is identified from an information model. 
     
     
         5 . The system of  claim 4 , wherein the information model is configured to generate the marker based on a combination indicators comprising different laboratory results. 
     
     
         6 . The system of  claim 4 , wherein the information model is configured to generate the marker based on a combination of vital signs and textual indicators, the vital signs and textual indicators being found in different documents the plurality of electronic documents. 
     
     
         7 . The system of  claim 4 , wherein the information model is configured to generate the marker based on a combination of indicators comprising structured and unstructured data in the medical record. 
     
     
         8 . The system of  claim 1 , wherein the query includes evidence describing at least one of the clinical indicators and associated contextual information from which the marker was generated. 
     
     
         9 . The system of  claim 1 , further comprising the computer system receiving confirmation of the query from the service provider. 
     
     
         10 . The system of  claim 1 , further comprising generating a confidence level for the marker based on the scenarios. 
     
     
         11 . The system of  claim 10 , further comprising receiving a selection from the service provider. 
     
     
         12 . The system of  claim 1 , wherein the multiple segments of text within the plurality of electronic documents comprise free text, a laboratory test result in combination with a treatment code, or a medical treatment code. 
     
     
         13 . A computer-based, natural language processing system comprising:
 a computer processor;   a database in communication with the computer processor, the database comprising memory for storing a plurality of electronic documents relating to patient care of a patient; and   a natural language processor module executing on the computer processor, the natural language processor module configured to:
 review the plurality of electronic documents to identify clinical indicators and associated contextual information by synthesizing descriptions of historical data using multiple segments of text within the plurality of electronic documents, wherein the clinical indicators and associated contextual information are analyzed according to a document type in which the clinical indicators are identified and according to the associated contextual factors related to the clinical indicator; 
 compare the clinical indicators and associated contextual information to rule-based scenarios; 
 generate a marker based on the clinical indicators and associated contextual information, the marker identifying a record element associated with a medical condition relating to the patient care; 
 generate a query for a service provider, the query describing the marker; and 
 verify a response to the query by the service provider by accessing the plurality of electronic documents to determine whether the service provider response to the query includes an appropriate amendment, correction, or update. 
   
     
     
         14 . The system of  claim 13 , further comprising generating a confidence level for the marker, based on the scenarios. 
     
     
         15 . The system of  claim 13 , wherein the multiple segments of text within the plurality of electronic documents comprise unstructured data, and wherein at least one of the clinical indicators is generated from the unstructured data. 
     
     
         16 . The system of  claim 15 , wherein the unstructured data comprises free text. 
     
     
         17 . The system of  claim 13 , wherein the multiple segments of text within the plurality of electronic documents comprise structured data, and wherein at least one of the clinical indicators is generated from the structured data. 
     
     
         18 . The system of  claim 17 , wherein the structured data comprise a laboratory test result in combination with a treatment code or a medical treatment code. 
     
     
         19 . The system of  claim 13 , further comprising verification presence of the indication of the medical condition in the medical record, based on the verified response by the service provider. 
     
     
         20 . The system of  claim 13 , wherein the query includes evidence describing at least one of the clinical indicators and associated contextual information from which the marker was generated.