Systems and methods for predicting characteristics of an artificial heart using an artificial neural network

1. A system configured to predict characteristics of an artificial heart, comprising:

a processor;

memory in electronic communication with the processor and

an artificial neural network in communication with the artificial heart, the artificial neural network being configured to:

receive an input vector of a predetermined length to train the artificial neural network;

produce an output vector based on the input vector;

compare the output vector with a target vector of the predetermined length;

when the output vector does not match the target vector within a predetermined error rate, adjust at least one weight;

when the output vector matches the target vector within the predetermined error rate, execute the input vector to produce an estimate of at least one of a flow rate, a differential pressure, a property relating to hematocrit and an orientation of a patient.

2. The system of claim 1 , wherein the artificial heart comprises an artificial heart having at least one magnetically levitated component.

3. The system of claim 1 , wherein the flow rate comprises positive fluid flow produced by the artificial heart.

4. The system of claim 1 , wherein the flow rate comprises negative fluid flow through the artificial heart.

5. The system of claim 1 , wherein the differential pressure comprises differential pressure exhibited across an inlet and an outlet of the artificial heart.

6. The system of claim 1 , wherein the artificial neural network is further configured to predict a condition of a patient that uses the artificial heart.

7. The system of claim 1 , wherein the input vector comprises at least one of rotor speed, rotor position, or motor current of the artificial heart.

8. The system of claim 1 , wherein the input vector comprises at least one of levitation current, system current, or system voltage.

9. The system of claim 1 , wherein the artificial heart comprises a left ventricular assist device (LVAD).

10. The system of claim 1 , wherein the artificial heart comprises a right ventricular assist device (RVAD).

11. The system of claim 1 , wherein the neural network is further configured to produce an estimate of a pulsitility index.

12. A computer-implemented method to predict characteristics of an artificial heart using an artificial neural network, comprising:

receiving, at a computer, an input vector of a predetermined length to train the artificial neural network, the artificial neural network being in communication wits the artificial heart;

producing, by the artificial neural network, an output vector based on the input vector;

comparing, by the computer, the output vector with a target vector of the predetermined length;

when the output vector does not match the target vector within a predetermined error rate, adjusting at least one weight of the artificial neural network;

when the output vector matches the target vector within the predetermined error rate, executing, by the artificial neural network, the input vector to produce an estimate of at least one of a flow rate, a differential pressure, a property relating to hematocrit and an orientation of a patient.

13. The method of claim 12 , wherein producing an estimate comprises producing an estimate if at least one characteristic of an artificial heart that comprise at least one magnetically levitated component.

14. The method of claim 12 , wherein the flow rate comprises positive fluid flow produced by the artificial heart.

15. The method of claim 12 , wherein the flow rate comprises negative fluid flow through the artificial heart.

16. The method of claim 12 , wherein the differential pressure comprises differential pressure exhibited across an inlet and an outlet of the artificial heart.

17. The method of claim 12 , further comprising using the artificial neural network to predict a condition of a patient that uses the artificial heart.

18. The method of claim 12 , wherein the input vector comprises at least one of rotor speed, rotor position, or motor current of the artificial heart.

19. The method of claim 12 , wherein the input vector comprises at least one of levitation current, system current, or system voltage.

20. The method of claim 12 , wherein the neural network is further configured to produce an estimate of a pulsitility index.

21. A computer-program product for predicting characteristics of an artificial heart using an artificial neural network, the computer-program product comprising a non-transitory computer-readable medium having instructions thereon, the instructions comprising:

code programmed to receive an input vector of a predetermined length to train the artificial neural network, the artificial neural network being in communication with the artificial heart;

code programmed to produce an output vector based on the input vector;

code programmed to compare the output vector with a target vector of the predetermined length;

when the output vector does not match the target vector within a predetermined error rate, code programmed to adjust at least one weight;

when the output vector matches the target vector within the predetermined error rate, code programmed to execute the input vector to produce an estimate of at least one of a flow rate a differential pressure, a property relating to hematocrit and an orientation of a patient.

22. The computer-program product of claim 21 , wherein the artificial heart comprises an artificial heart haling at least one magnetically levitated component.

23. The computer-program product of claim 21 , wherein the artificial heart comprises a left ventricular assist device (LVAD).

24. The computer-program product of claim 21 , wherein the artificial heart comprises a right ventricular assist device (RVAD).

25. A system configured to predict condition of a patient using an artificial heart, comprising:

a processor;

memory in electronic communication with the processor; and

an artificial neural network in communication with the artificial heart, the artificial neural network being configure to:

receive an input vector of a predetermined length to train the artificial neural network;

produce an output vector based on the input vector;

compare the output vector with a target vector of the predetermined length;

when the output vector does not match the target vector within a predetermined error rate, adjust at last one weight;

when the output vector matches the target vector within the predetermined error rate, execute the input vector to produce an estimate of at least one of a flow rate, a differential pressure, a property relating to hematocrit and an orientation of a patient.

26. The system of claim 25 , wherein the patient uses a natural heart in conjunction with the artificial heart.

27. The system of claim 25 , wherein the neural network is further configured to produce an estimate of a pulsitility index of the blood of the patient.

28. The system of claim 27 , wherein the pulsitility index describes the strength of a natural heart of the patient being used in conjunction with the artificial heart.

29. The system of claim 25 , wherein the neural network is further configured to produce an estimate of blood viscosity properties of the patient.

30. The system of claim 25 , wherein the neural network is further configured to produce an estimate of a recovery measurement of the natural heart of the patient.

31. The system of claim 30 , wherein the recovery measurement comprises a measurement of the contractility of a native ventricle of the natural heart of the patient.

32. The system of claim 30 , wherein the recovery measurement comprises a measurement of the elastance of a native ventricle of the natural heart of the patient.