EP-4736173-A1 - MACHINE LEARNING FOR REAL-TIME DECISION SUPPORT

EP4736173A1EP 4736173 A1EP4736173 A1EP 4736173A1EP-4736173-A1

Abstract

A computer system, computer program product and method for determining a probability of attaining a PK-PD target associated with efficacy for a patient that includes a processor(s) determining that a trigger event related to a patient, wherein the patient record comprises an order for a current drug regimen, has occurred in an electronic health record (EHR) system communicatively coupled to the one or more processors. The processor obtains descriptive information relating to a patient. The processor selects a pharmacokinetic model. The processor applies the pharmacokinetic model and utilizes the information relating to the patient to determine, for each of the one or more drug therapies, a probability of attaining a PK-PD target associated with efficacy for the patient with the infection. The processor generates rankings for the drug therapies and determines if the current drug regimen comprises a probability above a threshold. The processor generates a new order.

Inventors

AMBROSE, PAUL G.
BHAVNANI, Sujata M.
GILBERT, MICHAEL
RUBINO, CHRISTOPHER M.

Assignees

PRXCISION, INC.

Dates

Publication Date: 20260506
Application Date: 20240625

Claims (20)

1. A computer-implemented method comprising: determining, by one or more processors, that a trigger event related to a patient, wherein the patient record comprises an order for a current drug regimen, has occurred in an electronic health record (EHR) system communicatively coupled to the one or more processors; based on obtaining the notification, obtaining, by the one or more processors, from one or more electronic medical records (EMRs) stored in the EHR system communicatively coupled to the one or more processors, descriptive information relating to a patient, the descriptive information comprising one or more data elements selected from the group consisting of: an infection acquired by the patient, a pathogen isolated from the patient, a creatinine clearance of the patient, a weight of the patient, and a height of the patient; based on the one or more drug therapies, selecting a pharmacokinetic model; applying, by the one or more processors, the pharmacokinetic model and utilizing the information relating to the patient to determine, for each of the one or more drug therapies, a probability of attaining a PK-PD target associated with efficacy for the patient with the infection; automatically generating, by the one or more processors, rankings, for each of the one or more drug therapies, by ordering each probability of attaining the PK-PD target associated with efficacy for the patient with the infection, for each of the one or more drug therapies, for the one or more drug therapies, wherein the rankings comprise a ranked list with the probability of attaining a PK-PD target associated with efficacy for the patient with the infection for each of the one or more drug therapies, ranked in order of predicted efficacy; determining, by the one or more processors, based on the rankings, if the current drug regimen comprises a probability above a preconfigured threshold; and based on the determining and the rankings, generating a recommendation for a new order.
2. The computer-implemented method of claim 1, wherein the determining if the current drug regimen comprises the probability above the preconfigured threshold comprises determining that the current drug regimen comprises the probability above the preconfigured threshold, and wherein the generating comprises: generating, by the one or more processors, a clinical note of a recommendation, wherein the clinical note of a recommendation comprises the new order, and wherein the new order and is an order for the current drug regimen; and transmitting, by the one or more processors, the clinical node of the recommendation to the EHR system communicatively coupled to the one or more processors.
3. The computer-implemented method of claim 1, wherein the determining if the current drug regimen comprises the probability above the preconfigured threshold comprises determining that the current drug regimen does not comprise a probability above the preconfigured threshold, and wherein the generating comprises: generating, by the one or more processors, a clinical note of a recommendation, wherein the clinical note of a recommendation comprises the new order; and transmitting, by the one or more processors, the clinical node of the recommendation to the EHR system communicatively coupled to the one or more processors.
4. The computer-implemented method of claim 3, wherein generating the recommendation for a new order comprises: generating, by the one or more processors, an unsigned medication order based on the ranked list, the unsigned medication order comprising the new order.
5. The computer- implemented method of claim 3, wherein generating the recommendation for a new order comprises: transmitting, by the one or more processors, an alert to at least one user; obtaining, by the one or more processors, a response to the alert, wherein the response comprises a selection of a designation of a drug therapy from the one or more drug therapies comprising the ranked list, and wherein the drug therapy designated comprises the new order; and generating, by the one or more processors, an unsigned medication order comprising the new order.
6. The computer-implemented method of claim 5, wherein transmitting the alert comprises: generating, by the one or more processors, a message comprising a link to launch a graphical user interface, wherein the one or more processors automatically display the rankings in the graphical user interface upon selection of the link by a user receiving the message; and transmitting, by the one or more processors, the message to at least one user pre-defined to receive the message, wherein the user contact information is saved in a database communicatively coupled to the one or more processors, and wherein the response the selection of the designation of a drug therapy is performed by the user in the graphical user interface.
7. The computer-implemented method of claim 5, wherein transmitting the alert comprises: displaying, by the one or more processors, the rankings, in a graphical user interface, wherein the response the selection of the designation of a drug therapy is performed by the user in the graphical user interface.
8. The computer-implemented method of claim 1, wherein the trigger event comprises an update to at least one field of at least one electronic medical record in the EHR system.
9. The computer-implemented method of claim 1, wherein the trigger event is selected from the group consisting of: entry of a prescription for a given antibiotic, reception of a culture with a given pathogen, and entry of data comprising additional information about an existing prescription.
10. The computer-implemented method of claim 1, further comprising: retaining by the one or more processors, the recommendation on a memory device; prompting, by the one or more processors, through a user interface, a user to provide data indicating an actual efficacy of the drug therapy as utilized by the patient with the infection at one or more predetermined intervals after obtaining the recommendation; and obtaining, by the one or more processors, responsive to the prompting, the data indicating the actual efficacy of the drug therapy.
11. The computer- implemented method of claim 1, further comprising: generating or updating, by the one or more processors, based on data comprising the data indicating the actual efficacy, a base model, wherein the base model describes a relationship between given patient response and PK-PD target attainment that accounts based on patientspecific response modifiers.
12. The computer-implemented method of claim 11, wherein the data further comprises data selected from the group consisting of: patient demographic data, clinical data, and laboratory data.
13. The computer- implemented method of claim 1, wherein the selecting the pharmacokinetic model comprises: for each of the one or more drug therapies, determining a class for a PK-PD index; based on determining that a drug therapy of the one or more drug therapies is in a first class, selecting a pharmacokinetic model, wherein applying the pharmacokinetic model comprises evaluating total drug exposure in a 24-hour period, for the drug therapy, to determine the probability of attaining a PK-PD target associated with efficacy for the patient with the infection; and based on determining that a drug therapy of the one or more drug therapies is in a second class, selecting a pharmacokinetic model, wherein applying the pharmacokinetic model comprises evaluating % time above MIC, for the drug therapy, to determine the probability of attaining a PK-PD target associated with efficacy for the patient with the infection.
14. The computer-implemented method of claim 11, further comprising: obtaining, by one or more processors, additional information identifying an infection; based on the additional information, generating and displaying, by the one or more processors, a second list comprising one or more pathogens consistent with the additional information; obtaining, by the one or more processors, a first indication designating at least one pathogen from the second list comprising one or more pathogens from the second list; based on at the obtaining of the least one pathogen from the second list, generating, by the one or more processors, a third list comprising one or more drug therapies utilized to treat the at least one pathogen; obtaining, by the one or more processors, descriptive information relating to a second patient, the descriptive information comprising one or more data elements selected from the group consisting of: an infection acquired by the second patient, a pathogen isolated from the second patient, a creatinine clearance of the second patient, a weight of the second patient, and a height of the second patient; based on the one or more drug therapies in the third list, selecting a given pharmacokinetic model; applying, by the one or more processors, the given pharmacokinetic model and utilizing the information relating to the second patient and the base model to determine, for each of the one or more drug therapies of the third list, a probability of attaining a PK-PD target associated with efficacy for the second patient with the infection; and automatically generating, by the one or more processors, current rankings, for each of the one or more drug therapies of the third list, by ordering each probability of attaining the PK- PD target associated with efficacy for the second patient with the infection, for each of the one or more drug therapies of the third list, for the one or more drug therapies of the third list, wherein the current rankings comprise a ranked list with the probability of attaining a PK-PD target associated with efficacy for the second patient with the infection for each of the one or more drug therapies of the third list, ranked in order of predicted efficacy.
15. The computer-implemented method of claim 11, wherein the patient-specific response modifiers are selected from the group consisting of: previous antibiotic use, age, and clearing organ function.
16. The computer-implemented method of claim 1, wherein determining that the trigger event has occurred comprises: monitoring, by the one or more processors, logging of changes to the EHR system; and determining, by the one or more processors, based on the logging, that the trigger event has occurred.
17. The computer-implemented method of claim 1 , wherein determining that the trigger event has occurred comprises: obtaining, by the one or more processors, from an application programming interface communicatively coupled with the EHR system, a notification that the trigger event has occurred.
18. A computer system comprising: a memory; and one or more processors in communications with the memory, wherein the computer system is configured to perform a method, the method comprising: determining, by the one or more processors, that a trigger event related to a patient, wherein the patient record comprises an order for a current drug regimen, has occurred in an electronic health record (EHR) system communicatively coupled to the one or more processors; based on obtaining the notification, obtaining, by the one or more processors, from one or more electronic medical records (EMRs) stored in the EHR system communicatively coupled to the one or more processors, descriptive information relating to a patient, the descriptive information comprising one or more data elements selected from the group consisting of: an infection acquired by the patient, a pathogen isolated from the patient, a creatinine clearance of the patient, a weight of the patient, and a height of the patient; based on the one or more drug therapies, selecting a pharmacokinetic model; applying, by the one or more processors, the pharmacokinetic model and utilizing the information relating to the patient to determine, for each of the one or more drug therapies, a probability of attaining a PK-PD target associated with efficacy for the patient with the infection; automatically generating, by the one or more processors, rankings, for each of the one or more drug therapies, by ordering each probability of attaining the PK-PD target associated with efficacy for the patient with the infection, for each of the one or more drug therapies, for the one or more drug therapies, wherein the rankings comprise a ranked list with the probability of attaining a PK-PD target associated with efficacy for the patient with the infection for each of the one or more drug therapies, ranked in order of predicted efficacy; determining, by the one or more processors, based on the rankings, if the current drug regimen comprises a probability above a preconfigured threshold; and based on the determining and the rankings, generating a recommendation for a new order.
19. A computer program product comprising: a computer readable storage medium readable by a processing circuit and storing instructions for execution by the processing circuit comprising one or more processors for performing a method comprising: determining, by the one or more processors, that a trigger event related to a patient, wherein the patient record comprises an order for a current drug regimen, has occurred in an electronic health record (EHR) system communicatively coupled to the one or more processors; based on obtaining the notification, obtaining, by the one or more processors, from one or more electronic medical records (EMRs) stored in the EHR system communicatively coupled to the one or more processors, descriptive information relating to a patient, the descriptive information comprising one or more data elements selected from the group consisting of: an infection acquired by the patient, a pathogen isolated from the patient, a creatinine clearance of the patient, a weight of the patient, and a height of the patient; based on the one or more drug therapies, selecting a pharmacokinetic model; applying, by the one or more processors, the phaimacokinetic model and utilizing the information relating to the patient to determine, for each of the one or more drug therapies, a probability of attaining a PK-PD target associated with efficacy for the patient with the infection; automatically generating, by the one or more processors, rankings, for each of the one or more drug therapies, by ordering each probability of attaining the PK-PD target associated with efficacy for the patient with the infection, for each of the one or more drug therapies, for the one or more drug therapies, wherein the rankings comprise a ranked list with the probability of attaining a PK-PD target associated with efficacy for the patient with the infection for each of the one or more drug therapies, ranked in order of predicted efficacy; determining, by the one or more processors, based on the rankings, if the current drug regimen comprises a probability above a preconfigured threshold; and based on the determining and the rankings, generating a recommendation for a new order.
20. A computer-implemented method comprising: monitoring, by the one or more processors, an electronic health record (EHR) system communicatively coupled to the one or more processors; determining, by the one or more processors, based on the listening, that an event has occurred; determining, by the one or more processors, based on pre-configured logic, that the event comprises a covered event; based on determining that the event is a covered event, obtaining, by the one or more processors, values to generate a simulation to determine a response to the covered event, wherein the obtaining each value of the values is selected from the group consisting of: obtaining a value from a data source and deriving the value based on obtaining one or more parameters; simulating, by the one or more processors, based on the values a model, wherein the simulating results in a binary result indicating whether to proceed with an action or not to proceed with the action; and based on the binary result indicating to proceed with the action, transmitting, by the one or more processors, a recommendation to proceed with the action.

Description

MACHINE LEARNING FOR REAL-TIME DECISION SUPPORT CROSS REFERENCE TO RELATED APPLICATIONS [0001] This application claims priority to U.S. Provisional Patent Application Number 63/510,950, entitled “SYSTEM AND METHOD FOR RANKING OPTIONS FOR MEDICAL TREATMENTS,” which was filed on June 29, 2023. This application is also a continuation-in- part of U.S. Application No. 17/575,905, entitled “SYSTEM AND METHOD FOR RANKING OPTIONS FOR MEDICAL TREATMENTS,” which was filed on January 14, 2022, which is a continuation-in-part of U.S. Application No. 16/740,913, entitled “SYSTEM AND METHOD FOR RANKING OPTIONS FOR MEDICAL TREATMENTS,” filed January 13, 2020, which is a continuation-in-part of U.S. Application No. 14/600,948, entitled “SYSTEM AND METHOD FOR RANKING OPTIONS FOR MEDICAL TREATMENTS,” filed January 20, 2015. These applications are all hereby incorporated herein by reference in their entireties for all purposes. BACKGROUND OF INVENTION [0002] The invention relates generally to systems and methods for utilizing machine learning to provide real-time decision support, for example, to enable health care providers to discriminate among potential anti-infective therapies for the treatment of selected infectious diseases. [0003] Artificial intelligence (Al) refers to intelligence exhibited by machines. Artificial intelligence (Al) research includes search and mathematical optimization, neural networks, and probability. Artificial intelligence (Al) solutions involve features derived from research in a variety of different science and technology disciplines ranging from computer science, mathematics, psychology, linguistics, statistics, and neuroscience. Machine learning has been described as the field of study that gives computers the ability to learn without being explicitly programmed. [0004] The goal of anti-infective stewardship is to select therapies that optimize the probability of positive outcomes for patients suffering from an infection. The primary focus of anti-infective stewardship is the optimal selection of anti-infective therapy, including dose, dosing interval, and duration. Due to the emergence of anti-infective-resistant pathogens, selecting optimal anti-infective therapy is more complex than at any other time since the advent of penicillin. [0005] The correct therapy that optimizes the probability of a positive outcome can be impacted based on various changes being made within one or more interconnected electronic medical record systems. SUMMARY OF INVENTION [0006] Shortcomings of the prior art are overcome and additional advantages are provided through the provision of a method for providing real-time decision support for an electronic health record (EHR) system, the method includes: determining, by one or more processors, that a trigger event related to a patient, wherein the patient record comprises an order for a current drug regimen, has occurred in an electronic health record (EHR) system communicatively coupled to the one or more processors; based on obtaining the notification, obtaining, by the one or more processors, from one or more electronic medical records (EMRs) stored in the EHR system communicatively coupled to the one or more processors, descriptive information relating to a patient, the descriptive information comprising one or more data elements selected from the group consisting of: an infection acquired by the patient, a pathogen isolated from the patient, a creatinine clearance of the patient, a weight of the patient, and a height of the patient; based on the one or more drug therapies, selecting a pharmacokinetic model; applying, by the one or more processors, the pharmacokinetic model and utilizing the information relating to the patient to determine, for each of the one or more drug therapies, a probability of attaining a PK-PD target associated with efficacy for the patient with the infection; automatically generating, by the one or more processors, rankings, for each of the one or more drug therapies, by ordering each probability of attaining the PK-PD target associated with efficacy for the patient with the infection, for each of the one or more drug therapies, for the one or more drug therapies, wherein the rankings comprise a ranked list with the probability of attaining a PK-PD target associated with efficacy for the patient with the infection for each of the one or more drug therapies, ranked in order of predicted efficacy; determining, by the one or more processors, based on the rankings, if the current drug regimen comprises a probability above a preconfigured threshold; and based on the determining and the rankings, generating a recommendation for a new order. [0007] Shortcomings of the prior art are overcome, and additional advantages are provided through the provision of a computer program product for providing real-time decision support for an EHR system, the method includes. The computer program product comprises a storage medium readable by a one or more p