Application US20190156933

SYSTEMS AND METHODS FOR FULL BODY CIRCULATION AND DRUG CONCENTRATION PREDICTION
[+] Show Details

US20190156933

1. A system for predicting drug concentration levels as a function of time in one or more organs of a subject, comprising: (7) (3)

a subject database including at least one subject characteristic of the subject;

a drug database including at least one drug characteristic of a drug to be administered to the subject; and

a drug concentration prediction engine, the drug concentration engine configured to: execute a full body circulation engine using the at least one subject characteristic to determine a plurality of blood flow rates to the one or more organs of the subject;

determine a first concentration of the drug in a first blood flow entering a first organ based on an initial drug dosage and the blood flow rate to the first organ;

modify, using the at least one subject characteristic, a 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 a second organ downstream of the first organ based on the second concentration and the blood flow rate to the second organ;

modify, using the at least one subject characteristic, a second drug concentration prediction model of the second organ; and

determine a fourth concentration of the drug in the second organ using the modified second drug concentration prediction model.

2. The system of claim 1, wherein the drug concentration prediction engine is configured to output a first value at a first point in time of the second concentration of the drug, and output a second value at the first point in time of the fourth concentration of the drug. (0)

3. The system of claim 1, further comprising a drug concentration prediction display engine configured to generate a visualization of the second concentration of the drug in the first organ and the fourth concentration of the drug in the second organ, and cause a display device to display the visualization. (0)

4. The system of claim 1, further comprising a toxicity calculator configured to compare the second concentration of the drug in the first organ to a first toxicity threshold associated with the first organ, and output an indication of toxicity responsive to determining that the second concentration exceeds the first toxicity threshold. (1) (0)

6. The system of claim 1, further comprising a dosage recommendation engine configured to generate a dosage recommendation for the subject based on the subject profile and at least one of the second concentration of the drug or the fourth concentration of the drug. (0)

7. The system of claim 1, further comprising a drug recommendation engine configured to compare an efficacy to a threshold efficacy, and generate a recommendation to administer the drug to the patient based on the subject profile and whether the efficacy exceeds the threshold efficacy. (0)

8. The system of claim 1, wherein the at least one subject characteristic includes at least one of an age, sex, height, weight, disease condition, secondary condition, or genetic condition of the subject. (0)

9. The system of claim 1, wherein the at least one drug characteristic includes at least one of a solubility, a fraction unbound, a lipophilicity, an effective molecular weight, or a pKa of the drug. (0)

10. A method for predicting drug concentration levels as a function of time in one or more organs of a subject, comprising: (7) (9)

receiving at least one subject characteristic of the subject;

receiving at least one drug characteristic of a drug to be administered to the subject;

executing a full body circulation engine using the at least one subject characteristic to determine a plurality of blood flow rates to the one or more organs of the subject;

determining a first concentration of the drug in a first blood flow entering a first organ based on an initial drug dosage and the blood flow rate to the first organ;

modifying a first drug concentration prediction model of the first organ using the at least one subject characteristic;

determining a second concentration of the drug in the first organ using the modified first drug concentration prediction model;

determining a third concentration of the drug in a second blood flow entering a second organ downstream of the first organ based on the second concentration and the blood flow rate to the second organ;

modifying a second drug concentration prediction model of the second organ using the at least one subject characteristic; and

determining a fourth concentration of the drug in the second organ using the modified second drug concentration prediction model.

11. The method of claim 10, further comprising outputting a first value at a first point in time of the second concentration of the drug, and outputting a second value at the first point in time of the fourth concentration of the drug. (0)

12. The method of claim 10, further comprising generating a visualization of the second concentration of the drug in the first organ and the fourth concentration of the drug in the second organ, and causing a display device to display the visualization. (0)

13. The method of claim 10, further comprising comparing the second concentration of the drug in the first organ to a first toxicity threshold associated with the first organ, and outputting an indication of toxicity responsive to determining that the second concentration exceeds the first toxicity threshold. (0)

14. The method of claim 10, further comprising generating a dosage recommendation for the subject based on the subject profile and at least one of the second concentration of the drug or the fourth concentration of the drug. (0)

15. The method of claim 10, further comprising comparing an efficacy to a threshold efficacy, and generating a recommendation to administer the drug to the patient based on the subject profile and whether the efficacy exceeds the threshold efficacy. (0)

16. The method of claim 10, wherein the at least one subject characteristic includes at least one of an age, sex, height, weight, disease condition, secondary condition, or genetic condition of the subject. (0)

17. The method of claim 10, wherein the at least one drug characteristic includes at least one of a solubility, a fraction unbound, a lipophilicity, an effective molecular weight, or a pKa of the drug. (0)

18. A system for training a drug concentration prediction model, comprising: (6) (2)

a training database storing a plurality of subject profiles, each subject profile including a known drug concentration and a corresponding at least one subject characteristic; and

a machine learning engine including a plurality of drug concentration prediction models and at least one machine learning engine, the machine learning engine configured to: execute the plurality of drug concentration prediction models to generate a plurality of predicted drug concentrations;

compare the plurality of predicted drug concentrations to a corresponding plurality of known drug concentrations to generate a comparison result;

compare the comparison result to a comparison condition;

responsive to the comparison result satisfying the comparison condition, output the plurality of drug concentration prediction models; and

responsive to the comparison result not satisfying the comparison condition, adjust at least one parameter of the drug concentration prediction models.

19. The system of claim 18, wherein a first subject profile of the plurality of subject profiles includes a first subject characteristic, and a second subject profile of the plurality of subject profiles includes a second subject characteristic different from the first subject characteristic. (0)

20. The system of claim 18, wherein each drug concentration prediction model is configured to receive an initial drug concentration at a first point in time and output an output drug concentration at a selected second point in time subsequent to the first point in time. (0)

21. The system of claim 18, wherein a first drug concentration prediction model of the plurality of drug concentration models includes an unknown parameter including at least one of a partition coefficient, a blood to plasma ratio, a clearance value, or an organ-specific parameter, and a value of the unknown parameter is unknown or has an accuracy less than an accuracy threshold prior to generation of the comparison result by the machine learning engine. (0)

22. The system of claim 18, wherein the at least one machine learning algorithm includes at least one of a neural network or a support vector machine (SVM). (1) (0)

24. The system of claim 18, wherein the machine learning engine adjust the at least one parameter by executing an optimization algorithm to reduce or minimize the comparison result. (0)

25. The system of claim 18, wherein a first drug concentration prediction model of the plurality of drug concentration prediction models corresponds to a first organ of a subject, a second drug concentration prediction model of the plurality of drug concentration prediction models corresponds to a second organ of the subject downstream of the first organ, and executing the second drug concentration prediction model to generate a second predicted drug concentration for the second organ includes using a first predicted drug concentration generated by the first drug concentration prediction model. (0)

26. A method for training a drug concentration prediction model, comprising: (4) (6)

receiving a plurality of subject profiles, each subject profile including a known drug concentration and a corresponding at least one subject characteristic;

executing, by a machine learning engine including a plurality of drug concentration prediction models and at least one machine learning engine, the plurality of drug concentration prediction models to generate a plurality of predicted drug concentrations;

comparing the plurality of predicted drug concentrations to a corresponding plurality of known drug concentrations to generate a comparison result;

comparing the comparison result to a comparison condition;

responsive to the comparison result satisfying the comparison condition, outputting the plurality of drug concentration prediction models; and

responsive to the comparison result not satisfying the comparison condition, adjusting at least one parameter of the drug concentration prediction models.

27. The method of claim 26, wherein a first subject profile of the plurality of subject profiles includes a first subject characteristic, and a second subject profile of the plurality of subject profiles includes a second subject characteristic different from the first subject characteristic. (0)

28. The method of claim 26, wherein each drug concentration prediction model is configured to receive an initial drug concentration at a first point in time and output an output drug concentration at a selected second point in time subsequent to the first point in time. (0)

29. The method of claim 26, wherein a first drug concentration prediction model of the plurality of drug concentration models includes a partition coefficient, and a value of the partition coefficient is unknown or has an accuracy less than an accuracy threshold prior to generation of the comparison result by the machine learning engine. (0)

30. The method of claim 26, wherein the at least one machine learning algorithm includes at least one of a neural network or a support vector machine (SVM). (0)

View Abstract and Specification Size

PDF with Images and Document Face >

Full Text Publication >

Application US20190156933

SYSTEMS AND METHODS FOR FULL BODY CIRCULATION AND DRUG CONCENTRATION PREDICTION

Patent Matrix® Search