Automated pharmacy translation engine for prescription medication instructions
A system, apparatus, and method for providing an automated translation engine for translating a prescription into a standardized format is generally disclosed. More particularly, embodiments described in this disclosure relate to a system, apparatus, and method for automatically receiving a prescription form, analyzing information provided on the prescription form, and translating the information into a standardized form that details medication information including medication administration directions that are easy for a reader to understand.
1. A computing system for automatically producing medication labels that include prescription information and instructions in a standardized format based on prescription information in non-standard medical facility prescription formats, the computing system comprising:
a prescription information receiving module stored on a memory of a pharmacy management server that, when executed by a processor of the pharmacy management server, causes the processor of the pharmacy management server to automatically receive prescription information for a pharmacy entered via a user interface of a remote medical facility computing device by a medical professional,
wherein the prescription information is transmitted to the pharmacy management server, and
wherein the prescription information defines a medical facility-specific prescription format;
an automatic prescription translation module stored on the memory of the pharmacy management server that, when executed by the processor of the pharmacy management server, causes the processor of the pharmacy management server, to execute an automatic prescription translation process that causes the processor of the pharmacy management server to:
(a) obtain, from a data dictionary storing a corpus of translation rules and token category definitions, a plurality of translation rules and token category definitions based on the prescription information and the medical facility-specific prescription format;
(b) parse the prescription information into initial output components based upon the plurality of token category definitions, wherein the token categories include at least one of (i) frequency, (ii) dosage, (iii) duration, (iv) strength, or (v) administration related attributes;
(c) in response to the parsing, calculate a confidence score for the prescription information based on a plurality of confidence factors associated with the prescription information,
wherein the confidence factors include: presence of mandatory elements in the initial output components, presence of directions for administering a medication in the initial output components, presence of erroneous elements in the initial output components, combined character length of the initial output components, and presence of secondary inferred value in the initial output components;
(d) in response to calculating the confidence score:
add inferred primary elements to the prescription information based on at least one confidence factor and a mandatory token list, or
replace secondary inferred values in the prescription information with inferred primary elements based on at least one confidence factor and an inferred token list, wherein replacing secondary inferred values in the prescription information includes converting strength attributes to dosage attributes based on a drug strength parameter and a drug strength unit of measure (UOM) parameter;
(e) in response to the parsing, and when an exit condition is not satisfied, tag a word included in the prescription information that satisfies a token category definition with a corresponding token;
(f) modify the prescription information by modifying the word tagged with the token according to a translation rule corresponding to the token, wherein the modified prescription information is included in a final instruction wrapper that includes a long SIG generated by assembling the components of the parsed prescription information;
(g) validate the modified prescription information by determining whether the modified prescription information includes any clinically significant errors, wherein the clinically significant errors include at least: typical route of administration for the medication not matching formulated route of administration in the modified prescription information, and strength and dosage in the modified prescription information not being equal conversions;
(h) in response to the validating, and when the exit condition is satisfied, cause the automatic prescription translation process to cease,
wherein the exit condition is satisfied if three or more directions for administering the medication are identified in the modified prescription information; and
(i) in response to the parsing, and when more than three directions for administering the medication are identified in the modified prescription information, flag the modified prescription information for review; and
a medication label module stored on the memory of a pharmacy computer system that, when executed by a processor of the pharmacy computer system, causes the processor of the pharmacy computer system to automatically provide a medication label that includes the final instructions wrapper to a printer linked to the pharmacy computer system, in response to completing the automatic prescription translation process.
2. The computing system of claim 1 , wherein the word includes at least one lowercase letter and the token is a lowercase letter token; and wherein the processor is configured to modify the word according to a pre-processing translation rule that modifies the at least one lowercase letter included in the word into an uppercase letter.
3. The computing system of claim 1 , wherein the word includes at least one extra space in-between a next word in the prescription information and the token is an extra space token; and wherein the processor is configured to modify the word according to a pre-processing translation rule that modifies the at least one extra space into a single space.
4. The computing system of claim 1 , wherein the word includes at least one specified word, phrase, number, or character prestored in the data storage unit and the token is a specified replacement token; and wherein the processor is configured to modify the word according to a pre-processing translation rule that modifies the at least one specified word, phrase, number, or character with a corresponding standard word, phrase, number, or character prestored in the data storage unit.
5. The computing system of claim 1 , wherein the processor is configured to: tag a word in the prescription information identified as belonging to a removable word token category with a removable word token; parse the prescription information and identify words in the prescription information tagged with the removable word token; and modify the prescription information by removing words in the prescription information identified as being tagged with the removable word token.
6. The computing system of claim 1 , wherein the processor is configured to: control a first computing module to: parse the prescription information; in response to the parsing, tag the word included in the prescription information that satisfies the token category definition with the corresponding token; control a second computing module to: modify the prescription information by modifying the word tagged with the token according to the translation rule corresponding to the token; and generate the final instruction wrapper including the modified prescription information.
7. The computing system of claim 1 , wherein the final instruction wrapper is in a standard format for printing on a label and includes administration instructions for administering a medication identified in the prescription information.
8. A computer-implemented method for automatically translating prescription information and automatically producing medication labels that include prescription information and instructions in a standardized format based on prescription information in non-standard medical facility prescription formats, the method comprising:
a prescription information receiving module that, when executed by a processor the pharmacy management server, causes the processor of the pharmacy management server to automatically receive prescription information for a pharmacy entered via a user interface of a remote medical facility computing device by a medical professional, wherein the prescription information is transmitted to the pharmacy management server, and wherein the prescription information defines a medical facility-specific prescription format;
an automatic prescription translation module that, when executed by the processor of the pharmacy management server, causes the processor of the pharmacy management server to execute an automatic prescription translation process that causes the processor of the pharmacy management server to:
(a) obtain, from a data dictionary storing a corpus of translation rules and token category definitions, a plurality of translation rules and token category definitions based on the prescription information and the medical-facility prescription specific format;
(b) parse the prescription information into initial output components based upon the plurality of and token category definitions, wherein the token categories include at least one of (i) frequency, (ii) dosage, (iii) duration, (iv) strength, or (v) administration related attributes;
(c) in response to the parsing, calculate a confidence score for the prescription information based on a plurality of confidence factors associated with the prescription information,
wherein the confidence factors include: presence of mandatory elements in the initial output components, presence of directions for administering a medication in the initial output components, presence of erroneous elements in the initial output components, combined character length of the initial output components, and presence of secondary inferred value in the initial output components;
(d) in response to calculating the confidence score:
add inferred primary elements to the prescription information based on at least one confidence factor and a mandatory token list, or
replace secondary inferred values in the prescription information with inferred primary elements based on at least one confidence factor and an inferred token list, wherein replacing secondary inferred values in the prescription information includes converting strength attributes to dosage attributes based on a drug strength parameter and a drug strength unit of measure (UOM) parameter;
(e) in response to the parsing, and when an exit condition is not satisfied, tag a word included in the prescription information that satisfies a token category definition with a corresponding token;
(f) modify the prescription information by modifying the word tagged with the token according to a translation rule corresponding to the token, wherein the modified prescription information is included in a final instruction wrapper that includes a long SIG generated by assembling the components of the parsed prescription information;
(g) validate the modified prescription information by determining whether the modified prescription information includes any clinically significant errors, wherein the clinically significant errors include at least: typical route of administration for the medication not matching formulated route of administration in the modified prescription information, and strength and dosage in the modified prescription information not being equal conversions;
(h) in response to the validating, and when the exit condition is satisfied, cause the automatic prescription translation process to cease,
wherein the exit condition is satisfied if three or more directions for administering the medication are identified in the modified prescription information; and
(i) in response to the parsing, and when more than three directions for administering the medication are identified in the modified prescription information, flag the modified prescription information for review; and
a medication label module stored on the memory of a pharmacy computer system that, when executed by a processor of the pharmacy computer system, causes the processor of the pharmacy computer system to automatically provide a medication label that includes the final instructions wrapper to a printer linked to the pharmacy computer system, in response to completing the automatic prescription translation process.
9. The method of claim 8 , wherein the word includes at least one lowercase letter and the token is a lowercase letter token; and wherein controlling the processor to modify the word includes modifying the word according to a pre-processing translation rule that modifies the at least one lowercase letter included in the word into an uppercase letter.
10. The method of claim 8 , wherein the word includes at least one extra space in-between a next word in the prescription information and the token is an extra space token; and wherein controlling the processor to modify the word includes modifying the word according to a pre-processing translation rule that modifies the at least one extra space into a single space.
11. The method of claim 8 , wherein the word includes at least one specified word, phrase, number, or character prestored in the data storage unit and the token is a specified replacement token; and wherein controlling the processor to modify the word includes modifying the word according to a pre-processing translation rule that modifies the at least one specified word, phrase, number, or character with a corresponding standard word, phrase, number, or character prestored in the data storage unit.
12. The method of claim 8 , further comprising controlling the processor to control the automated prescription translation process to: tag a word in the prescription information identified as belonging to a removable word token category with a removable word token; parse the prescription information and identify words in the prescription information tagged with the removable word token; and modify the prescription information by removing words in the prescription information identified as being tagged with the removable word token.
13. The method of claim 8 , further comprising controlling the processor to control the automated prescription translation process to assign a default score to the prescription information.
14. The method of claim 8 , wherein the exit condition is satisfied when a mandatory element corresponding to a medication identified in the prescription information is not also included in the prescription information.
15. The method of claim 8 , wherein controlling the processor to control the automated prescription translation engine module includes: controlling a first computing module to: parse the prescription information; in response to the parsing, tag the word included in the prescription information that satisfies the token category definition with the corresponding token; and controlling a second computing module to: modify the prescription information by modifying the word tagged with the token according to the translation rule corresponding to the token; and generate the final instruction wrapper including the modified prescription information.
16. The method of claim 8 , wherein the prescription information is based on information obtained from an optical character recognition process implemented on a prescription note at the remote computing device.
17. The method of claim 8 , wherein the final instruction wrapper is in a standard format for printing on a label and includes administration instructions for administering a medication identified in the prescription information.
18. A non-transitory computer-readable medium storing computer-readable instructions that, when executed by a processor, cause the processor to automatically produce medication labels that include prescription information and instructions in a standardized format based on prescription information in non-standard medical facility prescription formats, the computer-readable medium comprising:
a prescription information data receiving module that, when executed by a processor of the pharmacy management server, causes the processor of the pharmacy management server to automatically receive prescription information for a pharmacy entered via a user interface of a remote medical facility computing device by a medical professional, wherein the prescription information is transmitted to the pharmacy management server, and wherein the prescription information defines a medical facility-specific prescription format;
an automatic prescription translation module that, when executed by the processor of the pharmacy management server, causes the processor of the pharmacy management server, to execute an automatic prescription translation process that causes the processors of the pharmacy management server to:
(a) obtain, from a data dictionary storing a corpus of translation rules and token category definitions, a plurality of translation rules and token category definitions based on the prescription information and the medical facility-specific prescription format;
(b) parse the prescription information into parsed initial output components based upon the plurality of and token category definitions, wherein the token categories include at least one of (i) frequency, (ii) dosage, (iii) duration, (iv) strength, or (v) administration related attributes;
(c) in response to the parsing, calculate a confidence score for the prescription information based on a plurality of confidence factors associated with the prescription information,
wherein the confidence factors include: presence of mandatory elements in the initial output components, presence of directions for administering a medication in the initial output components, presence of erroneous elements in the initial output components, combined character length of the initial output components, and presence of secondary inferred value in the initial output components;
(d) in response to calculating the confidence score:
add inferred primary elements to the prescription information based on at least one confidence factor and a mandatory token list, or
replace secondary inferred values in the prescription information with inferred primary elements based on at least one confidence factor and an inferred token list, wherein replacing secondary inferred values in the prescription information includes converting strength attributes to dosage attributes based on a drug strength parameter and a drug strength unit of measure (UOM) parameter;
(e) in response to the parsing, and when an exit condition is not satisfied, tag a word included in the prescription information that satisfies a token category definition with a corresponding token;
(f) modify the prescription information by modifying the word tagged with the token according to a translation rule corresponding to the token, wherein the modified prescription information is included in a final instruction wrapper that includes a long SIG generated by assembling the components of the parsed prescription information;
(g) validate the modified prescription information by determining whether the modified prescription information includes any clinically significant errors, wherein the clinically significant errors include at least: typical route of administration for the medication not matching formulated route of administration in the modified prescription information, and strength and dosage in the modified prescription information not being equal conversions;
(h) in response to the validating, and when the exit condition is satisfied, cause the automatic prescription translation process to cease,
wherein the exit condition is satisfied if three or more directions for administering the medication are identified in the modified prescription information; and
(i) in response to the parsing, and when more than three directions for administering the medication are identified in the modified prescription information, flag the modified prescription information for review; and
a medication label module that, when executed by a processor of the pharmacy computer system, causes the processor of the pharmacy computer system to automatically provide a medication label that includes the final instructions wrapper to a printer linked to the pharmacy computer system, in response to completing the automatic prescription translation process.
19. The non-transitory computer-readable medium of claim 18 , wherein the word includes at least one lowercase letter and the token is a lowercase letter token, and wherein the processor is further configured to modify the word according to a pre-processing translation rule that modifies the at least one lowercase letter included in the word into an uppercase letter.
20. The non-transitory computer-readable medium of claim 18 , wherein further execution of the instructions causes the processor to:
tag a word in the prescription information identified as belonging to a removable word token category with a removable word token;
parse the prescription information and identify words in the prescription information tagged with the removable word token; and
modify the prescription information by removing words in the prescription information identified as being tagged with the removable word token.
21. A computing system for an automatic prescription translation process, the computing system comprising:
a printer;
one or more processors; and
one or more non-transitory memories, the one or more non-transitory memories having stored thereon computer-executable instructions that, when executed by the one or more processors, cause the computing system to:
receive prescription information for a pharmacy entered via a user interface of a remote medical facility computing device by a medical professional,
wherein the prescription information is transmitted to the pharmacy management server, and
wherein the prescription information defines a medical facility-specific prescription format;
obtain, from a data dictionary storing a corpus of translation rules and token category definitions, a plurality of translation rules and token category definitions based on (i) the prescription information and (ii) the medical facility-specific prescription format;
parse the prescription information into initial output components based upon the plurality of token category definitions, wherein the token categories include at least one of (i) frequency, (ii) dosage, (iii) duration, (iv) strength, or (v) administration related attributes;
in response to parsing the prescription information, calculate a confidence score for the prescription information based on a plurality of confidence factors associated with the prescription information,
wherein the confidence factors include: presence of mandatory elements in the initial output components, presence of directions for administering a medication in the initial output components, presence of erroneous elements in the initial output components, combined character length of the initial output components, and presence of secondary inferred value in the initial output components;
in response to calculating the confidence score:
add inferred primary elements to the prescription information based on at least one confidence factor and a mandatory token list, or
replace secondary inferred values in the prescription information with inferred primary elements based on at least one confidence factor and an inferred token list, wherein replacing secondary inferred values in the prescription information includes converting strength attributes to dosage attributes based on a drug strength parameter and a drug strength unit of measure (UOM) parameter;
in response to parsing the prescription information, and when an exit condition is not satisfied, tag a word included in the prescription information that satisfies a token category definition with a corresponding token;
modify the prescription information by modifying the word tagged with the token according to a translation rule corresponding to the token, wherein the modified prescription information is included in a final instruction wrapper that includes a long SIG generated by assembling the components of the parsed prescription information;
validate the modified prescription information by determining whether the modified prescription information includes any clinically significant errors, wherein the clinically significant errors include at least: typical route of administration for the medication not matching formulated route of administration in the modified prescription information, and strength and dosage in the modified prescription information not being equal conversions;
in response to validating the prescription information, and when the exit condition is satisfied, cause the automatic prescription translation process to cease,
wherein the exit condition is satisfied if three or more directions for administering the medication are identified in the modified prescription information; and
in response to parsing the prescription information, and when more than three directions for administering the medication are identified in the modified prescription information, flag the modified prescription information for review; and
transmit, to the printer, a medication label that includes the final instruction wrapper in response to completing the automatic prescription translation process.