Apparatus and method for medication delivery using single input-single output (SISO) model predictive control

• US 8,945,094 B2
• Filed: 09/08/2010
• Issued: 02/03/2015
• Est. Priority Date: 09/08/2010
• Status: Active Grant

• Current Assignee: Honeywell International Inc.
• Original Assignee: Honeywell International Inc.
• Inventors: Nordh, Pernell E. T.
• Primary Examiner(s): SCHMIDT, EMILY LOUISE
• Application Number:
  US12/877,795
• Publication Number:
  US 20120059351A1
• Time in Patent Office: 1,609 Days
• Field of Search: 604 65- 67, 604/503, 604/504, 604/890.1, 604/891.1
• US Class Current: 604/890.1
• CPC Class Codes: A61M 2205/3561 local, e.g. within room or hospital
• CPC Class Codes: A61M 2205/3592 using telemetric means, e.g. radio or optical transmission
• CPC Class Codes: A61M 2205/50 with microprocessors or computers
• CPC Class Codes: A61M 2205/502 User interfaces, e.g. screens or keyboards
• CPC Class Codes: A61M 2205/505 Touch-screens; Virtual keyboard or keypads; Virtual buttons; Soft keys; Mouse touches
• CPC Class Codes: A61M 2205/52 with memories providing a history of measured variating parameters of apparatus or patient
• CPC Class Codes: A61M 2230/06 Heartbeat rate only
• CPC Class Codes: A61M 2230/201 Glucose concentration
• CPC Class Codes: A61M 2230/30 Blood pressure A61M2230/04 takes precedence
• CPC Class Codes: A61M 5/14244 adapted to be carried by the patient, e.g. portable on the body
• CPC Class Codes: A61M 5/1723 using feedback of body parameters, e.g. blood-sugar, pressure measurement of body parameters A61B5/00
• CPC Class Codes: G16H 20/10 relating to drugs or medications, e.g. for ensuring correct administration to patients
• CPC Class Codes: G16H 50/50 for simulation or modelling of medical disorders

Abstract

A method includes receiving measurements from a sensor associated with a patient at a portable medication delivery device. The method also includes controlling delivery of medication to the patient at the portable medication delivery device using a single input, single output (SISO) model predictive control technique. The SISO model predictive control technique includes predicting a characteristic of the patient using the measurements and a model associated with the patient. The SISO model predictive control technique also includes determining whether the characteristic of the patient is predicted to fall outside of a desired range. In addition, the SISO model predictive control technique includes, if the characteristic of the patient is predicted to fall outside of the desired range, determining an amount of medication to deliver to the patient and delivering the determined amount of medication to the patient.

Claims

1. A method comprising:

receiving measurements from a sensor associated with a patient at a portable medication delivery device; and

controlling delivery of medication to the patient at the portable medication delivery device using a single input, single output (SISO) model predictive control technique that includes;

predicting a characteristic of the patient using the measurements and a model associated with the patient;

determining whether the characteristic of the patient is predicted to fall outside of a desired range; and

if the characteristic of the patient is predicted to fall outside of the desired range, determining an amount of medication to deliver to the patient and automatically delivering the determined amount of medication to the patient, wherein determining the amount of medication to deliver comprises predicting an effect the amount of medication will have on the characteristic of the patient based on the model and estimating the amount of medication to return the characteristic of the patient to the desired range using the predicted effect and the model;

after delivery of at least a portion of the amount of medication to the patient, determining how closely an actual effect the amount of medication has on the characteristic of the patient tracks with the predicted effect; and

in response to determining that a difference between the actual effect and the predicted effect exceeds a threshold, identifying a new model to use in predicting the characteristic of the patient,

wherein the amount of medication automatically delivered to the patient always satisfies one or more constraints associated with at least one of a maximum amount or a minimum amount of medication.

2. The method of claim 1, wherein:

a single input of the SISO model predictive control technique comprises the measurements from the sensor; and

a single output of the SISO model predictive control technique comprises a control signal for controlling a pump that delivers the medication to the patient.

3. The method of claim 1, wherein the model comprises a generic model associated with data from multiple patients.

4. The method of claim 3, further comprising:

replacing the generic model with a patient-specific model associated with the individual patient; and

controlling the delivery of medication to the patient using the patient-specific model.

5. The method of claim 1, wherein controlling the delivery of medication to the patient further comprises:

minimizing the amount of medication needed in order to prevent the characteristic of the patient from falling outside of the desired range; and

ensuring that no more than the maximum amount of medication is delivered to the patient within a predetermined time period.

6. The method of claim 1, further comprising:

receiving input from the patient, the input adjustably defining a performance ratio of the SISO model predictive control technique;

wherein the adjustably defined performance ratio defines a speed at which the model predictive control technique responds to changes in the measurements from the sensor.

7. The method of claim 1, wherein:

the measurements from the sensor comprise blood glucose levels of the patient; and

controlling the delivery of medication to the patient comprises controlling delivery of insulin to the patient.

8. An apparatus comprising:

at least one interface configured to receive measurements from a sensor associated with a patient; and

a controller configured to control delivery of medication to the patient using a single input, single output (SISO) model predictive control technique that includes;

predicting a characteristic of the patient using the measurements and a model associated with the patient;

determining whether the characteristic of the patient is predicted to fall outside of a desired range; and

if the characteristic of the patient is predicted to fall outside of the desired range, determining an amount of medication to deliver to the patient and automatically initiating delivery of the determined amount of medication to the patient, wherein the controller is configured to predict an effect the amount of medication will have on the characteristic of the patient based on the model and estimate the amount of medication to return the characteristic of the patient to the desired range using the predicted effect and the model;

after delivery of at least a portion of the amount of medication to the patient, determining how closely an actual effect the amount of medication has on the characteristic of the patient tracks with the predicted effect; and

in response to determining that a difference between the actual effect and the predicted effect exceeds a threshold, identifying a new model to use in predicting the characteristic of the patient,

wherein the controller is configured to determine the amount of medication to deliver to the patient such that the amount of medication automatically delivered to the patient always satisfies one or more constraints associated with at least one of a maximum amount or a minimum amount of medication.

9. The apparatus of claim 8, wherein:

a single input of the SISO model predictive control technique comprises the measurements from the sensor; and

a single output of the SISO model predictive control technique comprises a control signal for controlling a pump that delivers the medication to the patient.

10. The apparatus of claim 8, wherein the model comprises a generic model associated with data from multiple patients.

11. The apparatus of claim 10, wherein the controller is further configured to:

replace the generic model with a patient-specific model associated with the individual patient; and

control the delivery of medication to the patient using the patient-specific model.

12. The apparatus of claim 11, wherein the controller is further configured to generate the patient-specific model.

13. The apparatus of claim 8, wherein the controller is configured to control the delivery of medication to the patient further by:

minimizing the amount of medication needed in order to prevent the characteristic of the patient from falling outside of the desired range; and

ensuring that no more than the maximum amount of medication is delivered to the patient within a predetermined time period.

14. The apparatus of claim 8, wherein:

the controller is further configured to receive input from the patient, the received input adjustably defining a performance ratio of the SISO model predictive control technique; and

the adjustably defined performance ratio defines a speed at which the model predictive control technique responds to changes in the measurements from the sensor.

15. The apparatus of claim 8, wherein:

the measurements from the sensor comprise blood glucose levels of the patient; and

the controller is configured to control delivery of insulin to the patient.

16. A non-transitory computer readable medium embodying a computer program, the computer program comprising computer readable program code for:

receiving measurements from a sensor associated with a patient at a portable medication delivery device; and

controlling delivery of medication to the patient at the portable medication delivery device using a single input, single output (SISO) model predictive control technique that includes;

predicting a characteristic of the patient using the measurements and a model associated with the patient;

determining whether the characteristic of the patient is predicted to fall outside of a desired range; and

if the characteristic of the patient is predicted to fall outside of the desired range, determining an amount of medication to deliver to the patient and automatically delivering the determined amount of medication to the patient, wherein the computer readable program code for determining the amount of medication to deliver comprises computer readable program code for predicting an effect the amount of medication will have on the characteristic of the patient based on the model and estimating the amount of medication to return the characteristic of the patient to the desired range using the predicted effect and the model;

after delivery of at least a portion of the amount of medication to the patient, determining how closely an actual effect the amount of medication has on the characteristic of the patient tracks with the predicted effect; and

in response to determining that a difference between the actual effect and the predicted effect exceeds a threshold, identifying a new model to use in predicting the characteristic of the patient,

wherein the amount of medication automatically delivered to the patient always satisfies one or more constraints associated with at least one of a maximum amount or a minimum amount of medication.

17. The non-transitory computer readable medium of claim 16, wherein:

a single input of the SISO model predictive control technique comprises the measurements from the sensor; and

a single output of the SISO model predictive control technique comprises a control signal for controlling a pump that delivers the medication to the patient.

18. The non-transitory computer readable medium of claim 16, further comprising:

computer readable program code for receiving input from the patient, the input adjustably defining a performance ratio of the SISO model predictive control technique;

wherein the adjustably defined performance ratio defines a speed at which the model predictive control technique responds to changes in the measurements from the sensor.

19. The non-transitory computer readable medium of claim 16, wherein:

the measurements from the sensor comprise blood glucose levels of the patient; and

the computer readable program code for controlling the delivery of medication to the patient comprises computer readable program code for controlling delivery of insulin to the patient.

20. The non-transitory computer readable medium of claim 16, further comprising:

computer readable program code for replacing a generic model with a patient-specific model associated with the patient.