Systems and methods for standardizing a pharmacy prescription
The present disclosure generally relates to systems and methods for inferring a standardized medication prescription from a medication prescription. A standardization engine of a prescription standardization system may receive a medication prescription and a request to standardize said medication prescription. In some cases, the standardization engine may normalize the medication prescription. In addition, the standardization engine, utilizing a machine learning model, may parse the medication prescription into atomic medical information. The machine learning model may infer a standardized medication prescription from the atomic medical information. In some embodiments, an active learning engine may be communicatively coupled to the standardization system. Systems and methods described herein also relate to the active learning engine to generate training sample sets (e.g., synthetic prescriptions) and train the machine learning model of the prescription standardization system.
1 . A system comprising:
a memory to store specific computer-executable instructions; and
a processor in communication with the memory, wherein the processor is to execute the specific computer-executable instructions to at least:
receive, from a user device, a request for a standardized medication prescription;
obtain a raw medication prescription, wherein the raw medication prescription includes medication data, medication direction data, and patient data associated with the prescription;
normalize the raw medication prescription into a normalized raw medication prescription;
input the normalized raw prescription direction to a machine learning model;
parse, with the machine learning model, the normalized raw medication prescription into atomic medical information;
infer, with the machine learning model, a standardized medication prescription from the atomic medical information;
supplement, with the machine learning model, the standardized medication prescription, wherein supplementing comprises:
identifying a portion of the standardized medication prescription;
generating supplemental information related to the portion of the standardized medication prescription based on a knowledge database; and
revising the standardized medical prescription to include the supplemental information; and
transmit the supplemented standardized medication prescription to the user device.
2 . The system of claim 1 , wherein the processor is to execute further specific computer-executable instructions to at least train the machine learning model on a set of training samples, the set of training samples comprising synthetic prescriptions.
3 . The system of claim 1 , wherein the processor is to execute further specific computer-executable instructions to at least augment the standardized medication prescription with domain knowledge data.
4 . The system of claim 3 , wherein the domain knowledge data comprises historical pharmacy directions.
5 . The system of claim 1 , wherein to infer, the processor is to execute further specific computer-executable instructions to calculate a dosage calculation based on the atomic medical information.
6 . The system of claim 1 , wherein the standardized medication prescription includes at least one of a minimal quantity per dose, a maximum quantity per dose, a daily administration minimum frequency, a daily administration maximum frequency, a cadence, hours of administration, or an average daily administration frequency.
7 . The system of claim 1 , wherein to normalize the raw medication prescription, the processor is to execute further specific computer-executable instructions to at least:
normalize text of the raw medication prescription according to domain knowledge data; and
authenticate the raw medication prescription according to the domain knowledge.
8 . A computer-implemented method comprising:
obtaining a medication prescription, wherein the medication prescription includes medication data, medication direction data, and patient data associated with the prescription;
inputting the medication prescription to a machine learning model;
parsing, with the machine learning model, the medication prescription into atomic medical information;
inferring, with the machine learning model, a standardized medication prescription from the atomic medical information;
supplementing, with the machine learning model, the standardized medication prescription, wherein supplementing comprises:
identifying a portion of the standardized medication prescription;
generating supplemental information related to the portion of the standardized medication prescription based on a knowledge database; and
revising the standardized medical prescription to include the supplemental information; and
transmitting the supplemented standardized medication prescription to the user device.
9 . The computer-implemented method of claim 8 , further comprising training the machine learning model on a set of training samples, the set of training samples comprising synthetic prescriptions.
10 . The computer-implemented method of claim 8 , further comprising augmenting the standardized medication prescription with domain knowledge data.
11 . The computer-implemented method of claim 10 , wherein the domain knowledge data comprises historical pharmacy directions.
12 . The computer-implemented method of claim 8 , wherein inferring the standardized medication prescription further comprises calculating a dosage calculation based on the atomic medical information.
13 . The computer-implemented method of claim 8 , wherein the standardized medication prescription includes, at least one of a minimal quantity per dose, a maximum quantity per dose, a daily administration minimum frequency, a daily administration maximum frequency, a cadence, hours of administration, or an average daily administration frequency.
14 . The computer-implemented method of claim 8 , further comprising:
normalizing text of the medication prescription according to domain knowledge data; and
authenticating the medication prescription according to the domain knowledge data.