Back to results
US 10,854,326 B2Granted
Systems and methods for full body circulation and drug concentration prediction
AssigneeVerisim Life Inc.
InventorJyotika Varshney
Life of this patent
Estimating the term…
Abstract
A method for predicting drug concentration levels includes receiving at least one subject characteristic of the subject, and executing a full body circulation model by: determining a first concentration of the drug in a first blood flow entering a first organ determining a second concentration of the drug in the first organ, determining a third concentration of a drug in a third blood flow entering a second organ, the third blood flow downstream of the first organ, and determining, using the second organ model a fourth concentration of the drug in the second organ.
Claims16
A system for predicting drug concentration levels as a function of time in one or more organs of a subject, comprising:
- a subject database including a plurality of characteristics of the subject;
- a drug database including at least one characteristic of a drug to be administered to the subject;
- and
- one or more processors configured to:
- train a first drug concentration prediction model of a first organ by performing machine learning by iteratively performing the following steps until a first comparison result is minimized or is less than a first threshold difference:
- providing, to the first drug concentration prediction model, as input, a plurality of first subject profiles, each first subject profile corresponding to a respective first training subject, each first subject profile comprising i) an identifier of the drug, ii) a plurality of first known concentration-time pairs, each first known concentration-time pair comprising a first known concentration of the drug in the first organ at a first known time, and iii) a characteristic of the first training subject,
- outputting, by executing the first drug concentration prediction model, a plurality of first predicted concentration-time pairs, each first predicted concentration-time pair comprising a first predicted concentration of the drug at the first known time of a corresponding first known concentration-time pair,
- comparing each first predicted concentration of the drug to a respective first known concentration of the drug to generate the first comparison result;
- and
- adjusting at least one parameter of the first drug concentration prediction model based on the first comparison result;
- train a second drug concentration prediction model of a second organ by performing machine learning by iteratively performing the following steps until a second comparison result is minimized or is less than a second threshold difference:
- providing, to the second drug concentration prediction model, as input, a plurality of second subject profiles, each second subject profile corresponding to a respective second training subject, each second subject profile comprising i) an identifier of the drug, ii) a plurality of second known concentration-time pairs, each second known concentration-time pair comprising a second known concentration of the drug at a second known time, and iii) a characteristic of the second training subject,
- outputting, by executing the second drug concentration prediction model, a plurality of second predicted concentration-time pairs, each second predicted concentration-time pair comprising a second predicted concentration of the drug at the second known time of a corresponding second known concentration-time pair;
- comparing each second predicted concentration of the drug to a respective second known concentration of the drug to generate the second comparison result, and
- adjusting at least one parameter of the second drug concentration prediction model based on the second comparison result;
- determine, using a first subset of the plurality of characteristics of the subject, a first blood flow rate to the first organ and a second blood flow rate to the second organ of the subject;
- determine a first concentration of the drug in a first blood flow entering the first organ based on an initial drug dosage and the first blood flow rate to the first organ;
- modify, using a second subset of the plurality of characteristics of the subject, the trained first drug concentration prediction model of the first organ;
- determine a second concentration of the drug in the first organ using the modified first drug concentration prediction model;
- determine a third concentration of the drug in a second blood flow entering the second organ downstream of the first organ based on the second concentration and the second blood flow rate to the second organ;
- modify, using a third subset of the plurality of characteristics of the subject, the trained second drug concentration prediction model of the second organ;
- determine a fourth concentration of the drug in the second organ using the modified second drug concentration prediction model;
- provide, as output, an indication of a recommendation for administering the drug for the subject responsive to comparing:
- (i) the second concentration of the drug in the first organ to a first maximum concentration for the first organ;
- ii) a first area under curve (AUC) based on first concentration levels of the drug within a first time duration in the first organ to a first maximum AUC for the first organ;
- (iii) the fourth concentration of the drug in the second organ to a second maximum concentration for the second organ;
- and
- iv) a second AUC based on second concentration levels of the drug within a second time duration in the second organ to a second maximum AUC for the second organ.
25 more elements
A method, comprising:
- training, by one or more processors, a first drug concentration prediction model of a first organ by performing machine learning by iteratively performing the following steps until a first comparison result is minimized or is less than a first threshold difference:
- providing, as input to the first drug concentration prediction model, a plurality of first subject profiles, each first subject profile corresponding to a respective first training subject, each first subject profile comprising i) an identifier of a drug, ii) a plurality of first known concentration-time pairs, each first known concentration-time pair comprising a first known concentration of the drug at a first known time, and iii) a characteristic of the first training subject,
- outputting, by executing the first drug concentration prediction model, a plurality of first predicted concentration-time pairs, each first predicted concentration-time pair comprising a first predicted concentration of the drug at the first known time of a corresponding first known concentration-time pair,
- comparing each first predicted concentration of the drug to a respective first known concentration of the drug to generate the first comparison result;
- and
- adjusting at least one parameter of the first drug concentration prediction model responsive to the first comparison result;
- training, by the one or more processors, a second drug concentration prediction model of a second organ by performing machine learning by iteratively performing the following steps until a second comparison result is minimized or is less than a second threshold difference:
- providing, as input to the second drug concentration prediction model, a plurality of second subject profiles, each second subject profile corresponding to a respective second training subject, each second subject profile comprising i) an identifier of the drug, ii) a plurality of second known concentration-time pairs, each second known concentration-time pair comprising a second known concentration of the drug at a second known time, and iii) a characteristic of the second training subject,
- outputting, by executing the second drug concentration prediction model, a plurality of second predicted concentration-time pairs, each second predicted concentration-time pair comprising a second predicted concentration of the drug at the second known time of a corresponding second known concentration-time pair,
- comparing each second predicted concentration of the drug to a respective second known concentration of the drug to generate the second comparison result;
- and
- adjusting at least one parameter of the second drug concentration prediction model responsive to the second comparison result;
- determining, by the one or more processors using a first subject of a plurality of characteristics of a subject, a first blood flow rate of a first blood flow to the first organ and a second blood flow rate of a second blood flow to the second organ;
- determining, by the one or more processors, a first concentration of the drug in the first blood flow;
- modifying, by the one or more processors using a second subset of the plurality of characteristics of the subject, the trained first drug concentration prediction model;
- determining, by the one or more processors, a second concentration of the drug in the first organ using the modified first drug concentration prediction model and the first concentration of the drug;
- determining, by the one or more processors, a third concentration of the drug in a second blood flow entering the second organ downstream of the first organ based on the second concentration and the second blood flow rate to the second organ;
- modifying, by the one or more processors using a third subset of the plurality of characteristics of the subject, the trained second drug concentration prediction model;
- determining, by the one or more processors, a fourth concentration of the drug in the second organ using the modified second drug concentration prediction model and the third concentration of the drug;
- and
- outputting an indication of a recommendation for administering the drug for the subject responsive to comparing:
- (i) the second concentration of the drug in the first organ to a first maximum concentration for the first organ;
- ii) a first area under curve (AUC) based on first concentration levels of the drug within a first time duration in the first organ to a first maximum AUC for the first organ;
- (iii) the fourth concentration of the drug in the second organ to a second maximum concentration for the second organ;
- and
- iv) a second AUC based on second concentration levels of the drug within a second time duration in the second organ to a second maximum AUC for the second organ.
23 more elements
Dependencies are inferred from the claim text.