Automated population of electronic records

• US 9,824,691 B1
• Filed: 06/02/2017
• Issued: 11/21/2017
• Est. Priority Date: 06/02/2017
• Status: Active Grant

• Current Assignee: Sorenson Ip Holdings, LLC
• Original Assignee: Sorenson Ip Holdings, LLC
• Inventors: Montero, Adam, Brooksby, Scot Lorin
• Primary Examiner(s): Yen, Eric
• Application Number:
  US15/612,644
• Time in Patent Office: 172 Days
• CPC Class Codes: G10L 15/1822 Parsing for meaning understanding
• CPC Class Codes: G10L 15/26 Speech to text systems G10L15/08 takes precedence
• CPC Class Codes: G10L 15/28 Constructional details of speech recognition systems
• CPC Class Codes: G10L 15/32 Multiple recognisers used in sequence or in parallel; Score combination systems therefor, e.g. voting systems
• CPC Class Codes: G10L 17/00 Speaker identification or verification techniques
• CPC Class Codes: G10L 2015/088 Word spotting
• CPC Class Codes: G16H 10/60 for patient-specific data, e.g. for electronic patient records
• CPC Class Codes: G16Z 99/00 Subject matter not provided for in other main groups of this subclass

Abstract

A computer-implemented method to populate an electronic record may include generating first transcript data of first audio of a first speaker during a conversation between the first speaker and a second speaker. The method may also include generating second transcript data of second audio of the second speaker during the conversation and identifying one or more words from the first transcript data as being a value for a record field based on the identified words corresponding to the record field and the one or more words being from the first transcript data and not being from the second transcript data. The method may further include providing the identified words to an electronic record database as a value for the record field of a user record of the first speaker.

Claims

1. A computer-implemented method to populate an electronic record based on communications between two people, the method comprising:

establishing a communication session between a first device and a second device such that first device audio is sent from the first device to the second device and second device audio is sent from the second device to the first device during the communication session;

receiving the first device audio as the first device audio is routed to the second device;

receiving the second device audio as the second device audio is routed to the first device;

generating first transcript data of the first device audio, the first transcript data being a transcription of the first device audio and including a plurality of first data segments, each of the plurality of first data segments including one or more words from the transcription of the first device audio;

generating second transcript data of the second device audio, the second transcript data being a transcription of the second device audio and including a plurality of second data segments, each of the plurality of second data segments including one or more words from the transcription of the second device audio;

identifying one of the plurality of second data segments that includes a field identification word that is associated with an identifier for a general record field that is to be populated using information from the first device audio and the second device audio during the communication session;

selecting one or more first data segments of the plurality of first data segments that occur after the identified second data segment as selected first data segments; and

identifying one or more of the one or more words from the selected first data segments as being a value for the general record field based on the identified words corresponding to the general record field.

2. The method of claim 1, further comprising:

obtaining a first user identifier associated with the first device based on data from the first device that is used to establish the communication session;

selecting a first electronic record type from a plurality of electronic record types based on the first user identifier;

matching the general record field with a first record field of the first electronic record type based on the general record field and the first record field being configured for analogous values;

providing the first user identifier to an electronic health record database that stores the first electronic record type; and

providing to the electronic health record database the identified words as a value for the first record field of a user record of the first electronic record type, the user record being associated with the first user identifier in the electronic health record database.

3. The method of claim 2, further comprising interweaving the first transcript data and the second transcript data to generate third transcript data such that a combined transcription included in the third transcript data is substantially in chronological order.

4. The method of claim 3, further comprising providing to the electronic health record database the third transcript data as a value of a second record field of the user record.

5. The method of claim 1, further comprising providing to the second device the field identification word for presentation of the field identification word by the second device.

6. The method of claim 5, wherein the identified second data segment is identified based on the identified second data segment including the field identification word and the identified second data segment occurring, in time, after the presentation of the field identification word by the second device.

7. One or more non-transitory media configured to store instructions that in response to being executed by one or more processors cause one or more systems to perform the method of claim 1.

8. A computer-implemented method to populate an electronic record based on communications between two people, the method comprising:

generating, using a processing system, first transcript data of first audio of a first speaker during a conversation between the first speaker and a second speaker, the first transcript data being a transcription of the first audio;

generating, using the processing system, second transcript data of second audio of the second speaker during the conversation, the second transcript data being a transcription of the second audio;

obtaining a first user identifier associated with the first speaker;

identifying a plurality of general record fields that are to be populated using information from the first audio and the second audio during the conversation, the plurality of general record fields being identified based on the first user identifier;

after identifying the plurality of general record fields, identifying, using the processing system, one or more words from the first transcript data as being a value for a record field based on the identified words corresponding to the record field and the one or more words being from the first transcript data and not being from the second transcript data, wherein the record field is one of the identified plurality of general record fields; and

providing, over a network, the identified words to an electronic record database as a value for the record field of a user record of the first speaker.

9. The method of claim 8, further comprising:

identifying one or more second words from the second transcript data as being a value for a second record field based on the identified second words corresponding to the second record field and the one or more second words being from the second transcript data and not the first transcript data; and

providing the identified second words to the electronic record database as a value for the second record field of a user record.

10. The method of claim 8, further comprising establishing a communication session between a first device and a second device such that first device audio is sent from the first device to the second device and second device audio is sent from the second device to the first device during the communication session, wherein the first audio is the first device audio and the second audio is the second device audio.

11. The method of claim 8, wherein the record field is a first record field and before providing the identified words the method further comprises:

selecting a first electronic record type from a plurality of electronic record types based on the first user identifier, the user record being of the first electronic record type; and

determining a second record field in the first electronic record type that corresponds to the first record field.

12. The method of claim 8, wherein the second speaker is a health care professional and the first speaker is a patient of the health care professional.

13. One or more non-transitory media configured to store instructions that in response to being executed by one or more processors cause one or more systems to perform the method of claim 8.

14. A computer-implemented method to populate an electronic record based on communications between two people, the method comprising:

establishing a communication session between a first device and a second device such that first device audio is sent from the first device to the second device and second device audio is sent from the second device to the first device during the communication session;

generating first transcript data of the first device audio, the first transcript data being a transcription of the first device audio;

generating second transcript data of the second device audio, the second transcript data being a transcription of the second device audio;

obtaining a first user identifier associated with the first device based on data from the first device that is used to establish the communication session;

identifying a plurality of general record fields that are to be populated using information from the first device audio and the second device audio during the communication session, the plurality of general record fields are identified based on the first user identifier;

after identifying the plurality of general record fields, identifying one or more words from the first transcript data as being a value for a record field based on the identified words corresponding to the record field and the one or more words being from the first transcript data and not being from the second transcript data, wherein the record field is one of the identified plurality of general record fields; and

providing the identified words to an electronic record database as a value for the record field of a user record.

15. The method of claim 14, wherein the record field is a first record field and before providing the identified words the method further comprises:

selecting a first electronic record type from a plurality of electronic record types based on the first user identifier, the user record being of the first electronic record type; and

determining a second record field in the first electronic record type that corresponds to the first record field.

16. The method of claim 14, further comprising:

identifying one or more second words from the second transcript data as being a value for a second record field based on the identified second words corresponding to the second record field and the one or more second words being from the second transcript data and not the first transcript data; and

providing the identified second words to the electronic record database as a value for the second record field of a user record.

17. One or more non-transitory media configured to store instructions that in response to being executed by one or more processors cause one or more systems to perform the method of claim 14.

18. A computer-implemented method to populate an electronic record based on communications between two people, the method comprising:

establishing a communication session between a first device and a second device such that first device audio is sent from the first device to the second device and second device audio is sent from the second device to the first device during the communication session;

generating first transcript data of the first device audio, the first transcript data being a transcription of the first device audio and including a plurality of first data segments, each of the plurality of first data segments including one or more words from the transcription of the first device audio;

generating second transcript data of the second device audio, the second transcript data being a transcription of the second device audio and including a plurality of second data segments, each of the plurality of second data segments including one or more words from the transcription of the second device audio;

identifying one of the plurality of second data segments that includes a field identification word that is associated with an identifier for a record field;

selecting an initial first data segment of the plurality of first data segments that occurs after the identified second data segment as a selected first data segment;

identifying one or more words from the first transcript data as being a value for the record field based on the identified words corresponding to the record field, being from the selected first data segment, and being from the first transcript data and not being from the second transcript data; and

providing the identified words to an electronic record database as a value for the record field of a user record.

19. The method of claim 18, further comprising:

identifying one or more second words from the second transcript data as being a value for a second record field based on the identified second words corresponding to the second record field and the one or more second words being from the second transcript data and not the first transcript data; and

providing the identified second words to the electronic record database as a value for the second record field of a user record.

20. One or more non-transitory media configured to store instructions that in response to being executed by one or more processors cause one or more systems to perform the method of claim 18.