Transcription data extraction

• US 7,613,610 B1
• Filed: 03/14/2005
• Issued: 11/03/2009
• Est. Priority Date: 03/14/2005
• Status: Active Grant

• Current Assignee: DeliverHealth Solutions LLC
• Original Assignee: eScription, Inc. (Microsoft Corporation)
• Inventors: Zimmerman, Roger S., Egerman, Paul, Zavaliagkos, George
• Primary Examiner(s): Chawan; Vijay B
• Application Number:
  US11/080,689
• Time in Patent Office: 1,695 Days
• Field of Search: 704/235, 704/9, 704/10, 704/2, 704/257, 704/260, 705/2, 705/3, 707/6, 707/5, 707/3, 379/88.01, 379/88.02
• US Class Current: 704/235
• CPC Class Codes: G10L 2015/228 of application context
• CPC Class Codes: G16H 10/60 for patient-specific data, e.g. for electronic patient records
• CPC Class Codes: Y10S 707/99935 Query augmenting and refining, e.g. inexact access
• CPC Class Codes: Y10S 707/99936 Pattern matching access

Abstract

A computer program product, for performing data determination from medical record transcriptions, resides on a computer-readable medium and includes computer-readable instructions for causing a computer to obtain a medical transcription of a dictation, the dictation being from medical personnel and concerning a patient, analyze the transcription for an indicating phrase associated with a type of data desired to be determined from the transcription, the type of desired data being relevant to medical records, determine whether data indicated by text disposed proximately to the indicating phrase is of the desired type, and store an indication of the data if the data is of the desired type.

Claims

1. At least one computer readable medium having stored thereon instructions that, when executed by at least one processor, perform a method of formatting a medical transcription transcribed by performing automatic speech recognition on a medical dictation, the method comprising:

receiving the medical transcription and a worktype code identifying a category type of the medical transcription;

selecting a first set of fields from a plurality of sets, the first set including a plurality of fields, each of the plurality of fields defining data of a particular type, each set of the plurality of sets associated with a respective category type of medical transcription, the first set being selected based, at least in part, on the worktype code;

automatically analyzing the medical transcription to identify at least one indicating phrase associated with at least one field in the first set;

determining whether text disposed proximately to the at least one indicating phrase is of the particular type corresponding to the at least one field; and

formatting the document by associating the text disposed proximately to the at least one indicating phrase with the at least field if the text is of the particular type corresponding to the at least one field such that, when the medical transcription is displayed, the text is displayed as data in the corresponding at least one field.

2. The at least one computer readable medium of claim 1, wherein the first set includes a plurality of fields, and wherein automatically analyzing includes, for each field of the plurality of fields in the first set, automatically analyzing the medical transcription to identify at least one indicating phrase associated with each field.

3. The at least one computer readable medium of claim 2, wherein associating comprises associating text determined to be of a particular type corresponding to any one of the plurality of fields such that, when the medical transcription is displayed, text identified as being the particular type is displayed as being part of the corresponding field.

4. The at least one computer readable medium of claim 3, wherein associating the text disposed proximately to the at least one indicating phrase includes adding the text as data to the corresponding field.

5. The at least one computer readable medium of claim 2, wherein associating the text disposed proximately to the at least one indicating phrase includes adding an indicator to the text disposed proximately to the at least one indicating phrase.

6. The at least one computer readable medium of claim 2, wherein each of the plurality of fields in the first set defines the particular type of data and has an associated label indicating the particular data type, and wherein formatting includes replacing the at least one indicating phrase with the associated label of the at least one of the plurality of fields.

7. The at least one computer readable medium of claim 6, wherein each of the plurality of fields in the first set for which no indicating phrase and/or text of the particular data type is found, presenting the field as vacant such that, when displayed, the associated label is displayed with an empty field to alert a user that no information was automatically identified for the vacant field.

8. The at least one computer readable medium of claim 1, wherein the text disposed proximately to the at least one indicating phrase is determined to be of the particular type only if a probability of the text being of the particular type is determined to exceed a threshold probability.

9. The at least one computer readable medium of claim 1, wherein the text disposed proximately to the at least one indicating phrase is determined to be of a first data type if a first probability that the text is of the first data type exceeds a second probability that the text is of a second data type.

10. The at least one computer readable medium of claim 1, further comprising analyzing information associated with a patient to whom the medical transcription pertains to assist in determining the particular type of data of the text disposed proximately the at least one indicating phrase based on known relationships between values of different data types and patient information.

11. The at least one computer readable medium of claim 10, further comprising automatically identifying the patient and/or automatically obtaining the patient information from the transcription.

12. The at least one computer readable medium of claim 1, further comprising automatically recognizing the medical transcription from the medical dictation using automatic speech recognition.

13. The at least one computer readable medium of claim 1, wherein the worktype code is automatically recognized from the medical dictation.

14. A system for formatting a medical transcription obtained from performing automatic speech recognition on a medical dictation, the system comprising:

at least one storage device including a data set that stores a plurality of fields, each of the plurality of fields defining data of a particular type and each of the plurality of fields belonging to one or more sets of a plurality of sets, each set of the plurality of sets associated with a respective category type of medical transcription;

at least one computer to receive the medical transcription and a work type code identifying a category type of the medical transcription, the at least one computer capable of accessing the database to select a first set from the plurality of sets based, at least in part, on the worktype code, the first set having a plurality of fields indicating a format for the medical transcription, the at least one computer programmed to implement;

a natural language processing module configured to automatically analyze the medical transcription to identify at least one indicating phrase associated with at least one of the plurality of fields and to determine whether text disposed proximately to the at least one indicating phrase is of the particular type corresponding to the at least one of the plurality of fields, the natural language processing module adapted to format the document by associating the text disposed proximately to the at least one indicating phrase with the at least one of the plurality of fields if the text is of the particular type corresponding to the at least one of the plurality of fields such that, when the medical transcription is displayed, the text is displayed as data in the corresponding at least one of the plurality of fields.

15. The system of claim 14, wherein the natural language processing module determines whether text disposed proximately to the at least one indicating phrase is of the particular type corresponding to the at least one of the plurality of fields based, at least in part, on a probabilistic determination.

16. The system of claim 15, wherein the probabilistic determination is based, at least in part, on patient information associated with a patient corresponding to the transcription.

17. The system of claim 14, wherein each field has an associated label indicating the particular type of data defined by the respective field, and wherein the natural language processing module formats the medical transcription by producing a table for display, the table including each of the plurality of fields in the first set and the associated label, wherein any text found to be of the particular type of data associated with any of the plurality of fields is inserted into the table in the corresponding field.

18. The system of claim 17, wherein the natural language processing module is configured to remove the at least one indicating phrase such that, when displayed, the table is displayed with the associated label replacing the at least one indicating phrase.

19. The system of claim 14, wherein the at least one indicating phrase comprises a natural language phrase.

20. The system of claim 14, wherein at least a portion of the transcription is normalized and the at least one indicating phrase comprises a normalized language phrase.

21. The system of claim 14, wherein the probabilistic determination includes comparing a first probability that the text is of the particular data type and a second probability that the text is of another data type.

22. The system of claim 14, wherein the probabilistic determination includes determining that a probability that the text is of the particular data type exceeds a probability threshold.