Natural language understanding of conversational sources

1. One or more non-transitory computer-readable media having computer-executable instructions embodied thereon that when executed, provide a method for enhanced natural language processing, the media comprising:

generating a transcript of a voice input in real-time for a current encounter with a patient;

generating and causing display of a graphical user interface that is populated in real-time with the transcript, the transcript displayed in a transcript area;

extracting a plurality of conditions from the transcript;

validating each of the plurality of conditions based on data in an electronic health record of the patient of the patient,

wherein the electronic health record is a longitudinal electronic health record of the patient including at least one prior encounter,

wherein the validating comprises comparing each of the plurality of conditions to the data in the electronic health record of the patient and identifying the extracted conditions that exist in the longitudinal electronic health record,

identifying, for each of the validated plurality of conditions, a plurality of concepts related to the condition, wherein:

each concept of the plurality of concepts corresponds to the condition, and

each of the plurality of concepts is associated with a temporal function to identify that a concept of the plurality of concepts is a present issue or a past issue;

classifying each of the validated plurality of conditions that are validated based on the data in the electronic health record associated with the patient into one or more of a plurality of classes, each of the plurality of classes being defined by an ontology corresponding to the condition; and

generating and causing display of the graphical user interface to be populated in real-time with the validated plurality of conditions sorted into the one or more of the plurality of classes, each condition being displayed in a classification area.

2. The media of claim 1 , wherein extracting the plurality of conditions from the transcript comprises parsing, by a natural language processor, unstructured voice data within the voice input.

3. The media of claim 1 , wherein validating each of the plurality of conditions that corresponds to data in the electronic health record comprises determining, by a processor, whether the condition that is extracted corresponds to structured data stored in the electronic health record that is associated with the patient.

4. The media of claim 1 , wherein the plurality of classes comprises one or more class types of clinical findings, symptoms, problems, observations, medications, or procedures.

5. The media of claim 4 , wherein the plurality of concepts related to the condition comprises one or more of:

a particular symptom that is associated with the condition;

a particular problem that is associated with the condition;

a particular medication that is specific to treatment of the condition; or

a particular procedure that is specific to treatment of the condition.

6. The media of claim 1 further comprising identifying, via a processor, a particular ontology that corresponds to the condition.

7. The media of claim 6 , wherein the plurality of concepts related to the condition are identified using the particular ontology that is identified as corresponding to the condition.

8. The one or more non-transitory computer-readable media of claim 1 , further comprising:

generating, using a scribe, an output, wherein generating the output includes generating the transcript;

identifying a discrepancy between the output from the scribe and the electronic health record of the patient; and

providing an error notification, wherein the notification includes each of a reason for the error, at least a portion of the one or more voice inputs related to the error, and at least a portion of data from one or more validation sources conflicting with the one or more voice inputs.

9. A computerized method comprising:

generating a transcript of a voice input in real-time for a current encounter with a patient;

generating and causing display of a graphical user interface that is populated in real-time with the transcript, the transcript displayed in a transcript area;

extracting a plurality of conditions from the transcript;

validating each of the plurality of conditions based on data in an electronic health record of the patient of the patient,

wherein the electronic health record is a longitudinal electronic health record of the patient including at least one prior encounter,

wherein the validating comprises comparing each of the plurality of conditions to the data in the electronic health record of the patient and identifying the extracted conditions that exist in the longitudinal electronic health record;

identifying, for each of the validated plurality of conditions, a plurality of concepts related to the condition, wherein:

each concept of the plurality of concepts corresponds to the condition, and

each of the plurality of concepts is associated with a temporal function to identify that a concept of the plurality of concepts is a present issue or a past issue;

classifying each of the validated plurality of conditions that are validated based on the date in the electronic health record associated with the patient into one or more of a plurality of classes, each of the plurality of classes being defined by an ontology corresponding to the condition; and

generating and causing display of the graphical user interface to be populated in real-time with the validated plurality of conditions sorted into the one or more of the plurality of classes, each condition being displayed in a classification area.

10. The method of claim 9 , wherein extracting the plurality of conditions from the transcript comprises parsing, by a natural language processor, unstructured voice data within the voice input.

11. The method of claim 9 , wherein validating each of the plurality of conditions that corresponds to data in the electronic health record comprises determining, by a processor, whether the condition that is that is extracted corresponds to structured data stored in the electronic health record that is associated with the patient.

12. The method of claim 9 , wherein the plurality of classes comprises one or more class types of clinical findings, symptoms, problems, observations, medications, or procedures.

13. The method of claim 12 , wherein the plurality of concepts related to the condition comprises one or more of:

a particular symptom that is associated with the condition;

a particular problem that is associated with the condition;

a particular medication that is specific to treatment of the condition; or

a particular procedure that is specific to treatment of the condition.

14. The method of claim 9 further comprising identifying, via a processor, a particular ontology that corresponds to the condition.

15. The method of claim 14 , wherein the plurality of concepts related to the condition are identified using the particular ontology that is identified as corresponding to the condition.

16. The method according to claim 9 , wherein in response to an output of the validating being an error, flagging the output of the validating with a visual indicator.

17. A system comprising:

one or more processors configured to:

generate a transcript of a voice input in real-time for a current encounter with a patient;

generate and cause display of a graphical user interface that is populated in real-time with the transcript, the transcript displayed in a transcript area;

extract a plurality of conditions from the transcript;

validate each of the plurality of conditions based on data in an electronic health record of the patient,

wherein the electronic health record is a longitudinal electronic health record of the patient including at least one prior encounter,

wherein the validating comprises comparing each of the plurality of conditions to the data in the electronic health record of the patient and identifying the extracted conditions that exist in the longitudinal electronic health record;

identify, for each of the validated plurality of conditions, a plurality of concepts related to the condition, wherein:

each concept of the plurality of concepts corresponds to the condition, and

each of the plurality of concepts is associated with a temporal function to identify that a concept of the plurality of concepts is a present issue or a past issue;

classify each of the validated plurality of conditions that are validated based on the data in the electronic health record associated with the patient into one or more of a plurality of classes, each of the plurality of classes being defined by an ontology corresponding to the condition; and

generate and cause display of the graphical user interface to be populated in real-time with the validated plurality of conditions sorted into the one or more of the plurality of classes, each condition being displayed in a classification area.

18. The system of claim 17 , wherein the plurality of classes comprises one or more class types of clinical findings, symptoms, problems, observations, medications, or procedures.

19. The system of claim 18 , wherein the plurality of concepts related to the condition comprises one or more of:

a particular symptom that is associated with the condition;

a particular problem that is associated with the condition;

a particular medication that is specific to treatment of the condition; or

a particular procedure that is specific to treatment of the condition.

20. The system of claim 17 further comprising identifying, via a processor, a particular ontology that corresponds to the condition.

21. The system of claim 20 , wherein the plurality of concepts related to the condition are identified using the particular ontology that is identified as corresponding to the condition.