Clinical data reconciliation as part of a report generation solution
An automated system updates electronic medical records (EMRs) based on dictated reports, without requiring manual data entry into on-screen forms. A dictated report is transcribed by an automatic speech recognizer, and facts are extracted from the report and stored in encoded form. Information from a patient's report is also stored in encoded form. The resulting encoded information from the report and EMR are reconciled with each other, and changes to be made to the EMR are identified based on the reconciliation. The identified changes are made to the EMR automatially, without requiring manual data entry into the EMR.
1. A method performed by at least one computer processor executing computer program instructions stored on at least one non-transitory computer-readable medium, wherein the computer program instructions are executable by the at least one computer processor to perform a method, the method comprising:
(A) receiving a spoken audio stream representing speech of a user;
(B) automatically transcribing the spoken audio stream using an automated speech recognizer to produce a transcript of the speech; and
(C) automatically updating an electronic medical record (EMR) stored in a non-transitory computer-readable medium, based on the transcript, the automatically updating comprising:
(C) (1) automatically extracting a fact, relating to a concept, from structured text in the transcript;
(C) (2) identifying, from among a plurality of discrete data elements in the EMR, a discrete data element in the EMR that relates to the concept;
(C) (3) automatically extracting a fact, relating to the concept, from the discrete data element;
(C) (4) generating, based on a value of the fact from the transcript and a value of the fact from the discrete data element, a reconciled fact, the generating comprising:
(C) (4) (a) determining that the value of the fact extracted from the transcript is correct;
(C) (4)b) determining that the value of the fact extracted from the discrete data element is incorrect; and
(C) (4) (c) in response to the determinations of (C) (4) (a) and (C) (4) (b), automatically identifying the reconciled fact as the fact extracted from the transcript; and
(C) (5) automatically updating the discrete data element in the EMR based on the reconciled fact, comprising automatically replacing the fact in the discrete data element with the fact extracted from the transcript.
2. The method of claim 1 , wherein the transcript includes plain text, and wherein (C) (1) comprises automatically extracting the fact from the plain text.
3. The method of claim 1 , further comprising:
(D) automatically updating free-form text contained in the EMR based on text contained in the transcript.
4. The method of claim 3 , wherein (D) comprises replacing the free-form text contained in the EMR with the text contained in the transcript.
5. The method of claim 3 , wherein (D) comprises storing, in the EMR, the text contained in the transcript.
6. The method of claim 1 , wherein (C) (5) further comprises:
(C) (5) (a) sending an EMR update message, indicating the reconciled fact, to an EMR system.
7. The method of claim 6 , wherein (C) (5) further comprises:
(C) (5) (b) at the EMR system, receiving the EMR update message;
(C) (5) (c) at the EMR system, automatically updating the EMR based on the EMR update message.
8. A non-transitory computer-readable medium containing computer program instructions, wherein the computer program instructions are executable by the at least one computer processor to perform a method, the method comprising:
(A) receiving a spoken audio stream representing speech of a user;
(B) automatically transcribing the spoken audio stream using an automated speech recognizer to produce a transcript of the speech; and
(C) automatically updating an electronic medical record (EMR) stored in a non-transitory computer-readable medium, based on the transcript, the automatically updating comprising:
(C) (1) automatically extracting a fact, relating to a concept, from structured text in the transcript;
(C) (2) identifying, from among a plurality of discrete data elements in the EMR, a discrete data element in the EMR that relates to the concept;
(C) (3) automatically extracting a fact, relating to the concept, from the discrete data element;
(C) (4) generating, based on a value of the fact from the transcript and a value of the fact from the discrete data element, a reconciled fact, the generating comprising:
(C) (4) (a) determining that the value of the fact extracted from the transcript is correct;
(C) (4) (b) determining that the value of the fact extracted from the discrete data element is incorrect; and
(C) (4) (c) in response to the determinations of (C) (4) (a) and (C) (4) (b), automatically identifying the reconciled fact as the fact extracted from the transcript; and
(C) (5) automatically updating the discrete data element in the EMR based on the reconciled fact, comprising automatically replacing the fact in the discrete data element with the fact extracted from the transcript.
9. The non-transitory computer-readable medium method of claim 8 , wherein the transcript includes plain text, and wherein (C) (1) comprises automatically extracting the fact from the plain text.
10. The non-transitory computer-readable medium of claim 8 , wherein the method further comprises:
(D) automatically updating free-form text contained in the EMR based on text contained in the transcript.
11. The non-transitory computer-readable medium of claim 10 , wherein (D) comprises replacing the free-form text contained in the EMR with the text contained in the transcript.
12. The non-transitory computer-readable medium of claim 10 , wherein (D) comprises storing, in the EMR, the text contained in the transcript.
13. The non-transitory computer-readable medium of claim 8 , wherein (C) (5) further comprises:
(C) (5) (a) sending an EMR update message, indicating the reconciled fact, to an EMR system.
14. The non-transitory computer-readable medium of claim 13 , wherein (C) (5) further comprises:
(C) (5) (b) at the EMR system, receiving the EMR update message;
(C) (5) (c) at the EMR system, automatically updating the EMR based on the EMR update message.