Clinically intelligent parsing

• US 9,864,838 B2
• Filed: 02/20/2008
• Issued: 01/09/2018
• Est. Priority Date: 02/20/2008
• Status: Active Grant

• Current Assignee: Medicomp Systems Incorporated
• Original Assignee: Medicomp Systems Incorporated
• Inventors: Goltra, Peter S.
• Primary Examiner(s): Corrielus, Jean M
• Application Number:
  US12/071,375
• Publication Number:
  US 20090210450A1
• Time in Patent Office: 3,611 Days
• Field of Search: 704 1, 704 9
• CPC Class Codes: G06Q 50/00 Information and communication technology [ICT] specially adapted for implementation of business processes of specific business sectors, e.g. utilities or tourism healthcare informatics G16H
• CPC Class Codes: G16H 15/00 ICT specially adapted for medical reports, e.g. generation or transmission thereof
• CPC Class Codes: G16H 40/20 for the management or administration of healthcare resources or facilities, e.g. managing hospital staff or surgery rooms
• CPC Class Codes: G16H 70/20 relating to practices or guidelines
• CPC Class Codes: G16Z 99/00 Subject matter not provided for in other main groups of this subclass

Abstract

A method, a device and a system for correlating medical information of a first format to medical information of a second format are provided. The method includes parsing an input sequence representing textual information into plural terms; searching a medical database to associate each term with a medical diagnosis; and translating each term into a coded phrase previously associated with the medical diagnosis in the medical database.

Claims

1. A device of correlating medical information of a first format to medical information of a second format, the device comprising:

a parsing means for parsing an input sequence representing textual information into plural terms;

a searching means for searching a medical database to associate each term with a medical diagnosis and an additional word memory for synonyms of each parsed term and selecting at least one synonym term having a degree of correlation with a respective medical diagnosis in the medical database that is above a predetermined threshold;

a translating means for translating each searched and selected term into a coded phrase wherein each coded phrase classifies a respective term in one of plural categories of medical findings associated with the medical diagnosis and stored in the medical database;

computing means for determining a point total for each medical diagnosis based on a total number of coded phrases associated with each medical diagnosis, and ranking the medical diagnoses based on respective point totals; and

output means for outputting at least one of the input sequence, the parsed terms, and the coded phrase to a display and prompting a user with additional medical findings which have not yet been entered into the medical database upon selection of a highest medical ranked diagnosis.

2. The device of claim 1, wherein the parsing means, the searching means and the translating means are implemented in separate software modules.

3. The device of claim 2, wherein the software modules are executable in a shared processor.

4. The device of claim 2, wherein each of the software modules is executable in a dedicated processor, each dedicated processor being dedicated to one of the software modules.

5. The device of claim 1, wherein the parsing means, the searching means and the translating means are implemented in a shared software module.

6. The device of claim 1, wherein the coded phrase is selected from the categories of medical findings in the medical database consisting of:

medical symptom, medical history, physical, medical test, and therapy.

7. A system correlating medical information of a first format to medical information of a second format, the system comprising:

a processor, the processor being able to execute a software module;

one or more data storage devices coupled to the processor, the data storage devices storing a medical vocabulary file, and a medical database;

one or more input devices connected to the processor for sending input to the processor; and

a display device connected to the processor for displaying an output of the processor;

wherein the software module is executed to;

access the medical vocabulary file and the medical database to parse an input sequence representing textual information into plural terms, access the medical database and perform a search to associate each parsed term with a medical diagnosis, which includes access an additional word memory for synonyms of each parsed term and selecting at least one synonym term having a degree of correlation with a respective medical diagnosis in the medical database that is above a predetermined threshold, translate each searched and selected term into a coded phrase, determine a point total for each medical diagnosis based on a total number of coded phrases associated with each medical diagnosis, rank the medical diagnoses based on respective point totals, provide each searched and selected term and the coded phrase to the output of the processor, and prompt a user with additional medical findings which have not yet been entered into the medical database upon selection of a highest medical ranked diagnosis, wherein each coded phrase classifies a respective term in one of plural categories of medical findings associated with the medical diagnosis and stored in the medical database.

8. The system of claim 7, wherein the one or more input devices comprise a microphone.

9. The system of claim 8, wherein a cursor of the display device is controlled by an input to the microphone.

10. The system of claim 8, wherein the system further comprises a voice recognition software module which is executable to transcribe input to the microphone into textual information.

11. The system of claim 7, wherein the coded phrase is selected from the categories of medical findings in the medical database consisting of:

medical symptom, medical history, physical, medical test, and therapy.

12. A method of correlating medical information of a first format to medical information of a second format in a system having an input device, at least one processor, a first memory, and a second memory, the method comprising:

receiving, in the input device, an input sequence representing textual information; and

in at least one first processor;

parsing the input sequence into plural terms;

searching a medical database to associate each term with a medical diagnosis, which includes searching an additional word memory for synonyms of each parsed term and selecting at least one synonym term having a degree of correlation with a respective term in the medical database that is above a predetermined threshold, wherein the searching the medical database includes;

identifying term groups, each term group including terms associated with at least one common diagnosis; and

associating each term in the term group with the at least one common diagnosis, wherein the term groups are identified by;

determining whether an a first intersection set between one or more diagnoses corresponding to the first of plural terms and one or more diagnoses corresponding to the second of the plural terms is an empty set;

if the first intersection set is a non-empty set, identifying the first term and the second term as part of one group; and

if the first intersection set is an empty set, identifying the first term and the second term as being in separate groups;

translating each searched and selected term into a coded phrase wherein each coded phrase classifies a respective term into one of plural categories of medical findings associated with the medical diagnosis and stored in the medical database;

outputting at least one of the input sequence, the parsed terms, and the coded phrase to at least one of a second processor and a display; and

in the at least one second processor;

determining a point total for each medical diagnosis based on a total number of coded phrases associated with each medical diagnosis;

ranking the medical diagnoses based on respective point totals; and

prompting a user with additional medical findings not yet entered upon selection of a highest medical ranked diagnosis.

13. The method of claim 12, wherein the parsing of the input sequence comprises:

sequentially processing the input sequence by;

identifying whether a word of input sequence is included in a medical vocabulary file;

determining whether the word is part of a phrase of the input sequence using the medical database; and

if the word is part of the phrase, including the word into the phrase to form a new term.

14. The method of claim 12, wherein the term groups are identified by:

sequentially processing each remaining term in the parsed input sequence by;

determining whether an intersection set between one or more diagnoses corresponding to the term and one or more diagnoses corresponding to a term group that immediately precedes the term, which is a second intersection set, is an empty set;

if the second intersection set is a non-empty set, including the term into the immediately preceding term group.

15. The method of claim 14, wherein:

if the second intersection set is an empty set, determining whether an intersection set between the one or more diagnoses corresponding to the term and one or more diagnoses corresponding to an existing term group, which is a third intersection set, is an empty set; and

if the third intersection set is a non-empty set, including the term into the existing term group; and

if the third intersection set is an empty set, identifying the term as part of a new term group.

16. The method of claim 12, wherein the parsing of the input sequence and the identifying of the term groups are performed concurrently.

17. The method of claim 12, wherein the parsing of the input sequence and the identifying of the term groups are performed concurrently with a nested loop, an outer loop of the nested loop executing the parsing of the input sequence, and an inner loop executing the identifying of the term groups.

18. The method of claim 12, comprising:

deleting any word that is included in an ignore word file from the input sequence before the parsing the input sequence.

19. The method of claim 18, wherein the ignore word file includes prepositions.

20. The method of claim 12, comprising:

determining alternative terms associated with a term by searching an additional word file.

21. The method of claim 20, wherein determining the alternative terms comprises:

selecting one or more of the alternative terms that have the strongest correlation with terms that immediately precede and succeed the term.

22. The method of claim 21, wherein a degree of correlation between two terms is determined by the size of intersection set between diagnoses which refer to the two terms in the medical database.

23. The method of claim 12, wherein the input sequence comprises:

a voice input that has been transcribed.

24. The method of claim 12, wherein the input sequence comprises:

a recording of a dictation.

25. The method of claim 12, wherein the input sequence comprises:

a live dictation.

26. The method of claim 12, wherein the input sequence representing textual information comprises:

one or more instructions.

27. The method of claim 26, wherein the one or more instructions comprise:

a first instruction indicating that content following the first instruction relates to a diagnosis.

28. The method of claim 27, comprising:

detecting the first instruction; and

translating the one or more terms in a term group immediately successive to the first instruction into coded phrases associated with the diagnosis indicated in the first instruction in the medical database.

29. The method of claim 26, wherein the one or more instructions comprise:

a second instruction indicating that content following the second instruction describes a medical problem previously described in the input sequence.

30. The method of claim 29, comprising:

detecting the second instruction;

identifying an existing term group relating to the medical problem indicated in the second instruction; and

identifying terms in a term group immediately successive to the second instruction and the identified existing term group as one group.

31. The method of claim 12, comprising:

outputting the translated phrases; and

selecting from a plurality of clinical protocol forms based on the output.

32. The method of claim 12, wherein the method comprises:

outputting the translated phrases; and

using the output as an input of a system which receives coded phrases as input.

33. A non-transitory computer readable medium having stored therein a program, which when executed causes a processor to perform a method of correlating medical information of a first format to medical information of a second format in a system having an input device, at least one processor, a first memory, and a second memory, the method comprising:

receiving, in the input device, an input sequence representing textual information; and

in at least one first processor;

parsing the input sequence into plural terms;

searching a medical database to associate each term with a medical diagnosis, which includes searching an additional word memory for synonyms of each parsed term and selecting at least one synonym term having a degree of correlation with a respective term in the medical database that is above a predetermined threshold, wherein the searching the medical database includes;

identifying term groups, each term group including terms associated with at least one common diagnosis; and

associating each term in the term group with the at least one common diagnosis, wherein the term groups are identified by;

determining whether an a first intersection set between one or more diagnoses corresponding to the first of plural terms and one or more diagnoses corresponding to the second of the plural terms is an empty set;

if the first intersection set is a non-empty set, identifying the first term and the second term as part of one group; and

if the first intersection set is an empty set, identifying the first term and the second term as being in separate groups;

translating each searched and selected term into a coded phrase wherein each coded phrase classifies a respective term into one of plural categories of medical findings associated with the medical diagnosis and stored in the medical database;

outputting at least one of the input sequence, the parsed terms, and the coded phrase to at least one of a second processor and a display; and

in the at least one second processor;

determining a point total for each medical diagnosis based on a total number of coded phrases associated with each medical diagnosis;

ranking the medical diagnoses based on respective point totals; and

prompting a user with additional medical findings not vet entered upon selection of a highest medical ranked diagnosis.