Maintaining multiple defined physiological zones using model predictive control

• US 9,700,708 B2
• Filed: 04/02/2013
• Issued: 07/11/2017
• Est. Priority Date: 10/12/2010
• Status: Active Grant

• Current Assignee: Sansum Diabetes Research Institute, Regents of the University of California (University of California)
• Original Assignee: Regents of the University of California (University of California)
• Inventors: Doyle, III, Francis J., Grosman, Benyamin, Jovanovic, Lois, Zisser, Howard, Dassau, Eyal
• Primary Examiner(s): Mendez, Manuel
• Application Number:
  US13/854,963
• Publication Number:
  US 20130231642A1
• Time in Patent Office: 1,561 Days
• Field of Search: 604 65- 67, 604131, 604151, 128DIG 1, 128 12, 128 13, 600301, 600365
• CPC Class Codes: A61M 31/002 Devices for releasing a drug at a continuous and controlled rate for a prolonged period of time artificial gland structures or devices A61F2/022; intra-uterine contraceptive devices A61F6/14; tampons for introducing into the vagina A61F13/20, A61L15/00; suppositories or bougies for intra-vaginal or intra-uterine application A61K9/02; physical forms of medicinal preparations for sustained or differential drug release A61K9/20, A61K9/50
• CPC Class Codes: A61P 3/10 for hyperglycaemia, e.g. antidiabetics
• CPC Class Codes: G01N 2333/62 Insulins
• CPC Class Codes: G01N 33/66 involving blood sugars, e.g. galactose
• CPC Class Codes: G01N 33/74 involving hormones or other non-cytokine intercellular protein regulatory factors such as growth factors, including receptors to hormones and growth factors
• CPC Class Codes: G16H 20/17 delivered via infusion or injection
• CPC Class Codes: G16H 50/50 for simulation or modelling of medical disorders
• CPC Class Codes: G16Z 99/00 Subject matter not provided for in other main groups of this subclass

Abstract

The disclosure relates to drug delivery and maintaining multiple defined physiological zones using model predictive control.

Claims

1. A drug delivery controller comprising an input contact, an output contact, and a multi-zone model predictive control (MPC) algorithm having a plurality of defined physiological zones, each with a different predefined control tuning, wherein the algorithm processes an input signal from a physiological sensor at the input contact to form an output signal to a drug infuser at the output contact.

2. The controller of claim 1 wherein the algorithm provides:

(a) a relatively conservative control action for a large deviation from a normal or desired zone, and

(b) a relative aggressive control action for within the normal or desired zone.

3. The controller of claim 1 comprising a logic scheme as shown in FIG. 5, wherein the algorithm provides physiological zones (G) and control weights (Q and R) according to the following schedule:

(i) R₁>

>

R_a, when G>

=G_m+1;

(ii) G₁ is an undesired zone that requires aggressive action that will result with suspension of delivery, wherein Q₁>

>

>

than Q_{2, 3, . . . , m};

(iii) desired zone, Q is zero;

(iv) for a large deviation from the normal or desired zone (G_m), Q_m<

Q<sub>m−

1</sub>, but not zero, and <

<

than Q₁;

wherein “

>

>

”

means at least one order of magnitude greater; and

“

>

>

>

”

means at least two orders of magnitude.

4. The controller of claim 1 comprising a logic scheme as shown in FIG. 5, wherein Q/R are defined as a, b, c and d for four physiological zones, and:

(i) a<

b;

(ii) b/a>

1×

10³;

(iii) c=0; and

(iv) d>

>

>

than a, b and c.

5. The controller of claim 1 comprising a logic scheme as shown in FIG. 5, wherein Q/R are defined as a, b, c and d for four physiological zones, and:

(i) a is 2×

10⁴;

(ii) b is 2;

(iii) c=0; and

(iv) d is 2×

10¹⁰.

6. The controller of claim 1 wherein the physiological zones are blood glucose concentration zones, the drug is insulin, and the algorithm provides:

(a) a relative conservative control action for a elevated “

hyperglycemia”

zone; and

(b) a relatively aggressive control action for a normal or near normal “

euglycemia”

zone.

7. The controller of claim 1 wherein the physiological zones are blood glucose concentration zones, the drug is insulin, and the algorithm provides:

(a) a relative conservative control action for a elevated “

hyperglycemia”

zone;

(b) a relatively aggressive control action for a normal or near normal “

euglycemia”

zone;

(c) control inaction for a normoglycemia zone; and

(d) a control action that results in suspending insulin delivery for a hypoglycemia zone.

8. The controller of claim 1 wherein the algorithm provides glycemia zones and control weights ratios (Q_k and R_k) according to the following schedule:

9. The controller of claim 1 further comprising the physiologic sensor and the drug infuser.

10. The controller of claim 1 further comprising the physiologic sensor, which is a glucose sensor, and the drug infuser, which is an insulin infuser.

11. A method of delivering a drug comprising:

(a) sensing a physiological metric with a sensor and receiving the resultant signal at the input contact of the controller of claim 1;

(b) processing the input signal to form the output signal;

(c) transmitting the output signal from the output contact to an drug infuser; and

(c) delivering the drug from the infuser pursuant to the output signal.

12. A method of delivering insulin comprising:

(a) sensing blood glucose concentration with a glucose sensor and receiving the resultant signal at the input contact of the controller of claim 1;

(b) processing the input signal to form the output signal;

(c) transmitting the output signal from the output contact to an insulin infuser; and

(d) delivering insulin from the infuser pursuant to the output signal.

13. A method of delivering a drug, comprising:

(a) obtaining drug and event data for a subject using an open loop protocol to obtain time values, drug values and physiological metric values;

(b) defining a plurality of physiological zones as a function of physiological metric values; and

(c) programming an automated drug pump comprising a controller having a multi-zone-model predictive control (MPC) algorithm,

wherein the controller obtains physiological metric measurements from the subject, compares the physiological metrics to the plurality of physiological zones and causes delivery of the drug to the subject pursuant to the MPC algorithm.

14. A method of claim 13 of delivering insulin, comprising:

(a) obtaining insulin and meal data for a subject using an open loop protocol to obtain time values, insulin values and glucose values;

(b) defining a plurality of glycemia zones as a function of blood glucose concentration; and

(c) programming an automated insulin pump comprising a controller having a multi-zone-model predictive control algorithm,

wherein the controller obtains glucose measurements from the blood, compares the glucose values to the plurality of glycemia zones and causes delivery of insulin to the subject.

15. The method of claim 14, wherein the plurality of zones is set according to the following schedule:

16. A method of delivering a drug comprising:

(a) determining physiological metric values from the blood of a subject;

(b) comparing the physiological metric values to a plurality of defined physiological zones defined as a function of the physiological metric; and

(c) automatically delivering the via a drug pump comprising a controller having a multi-zone-model predictive control (MPC) algorithm comprising the physiological zones.

17. A method of claim 16 of delivering insulin comprising:

(a) determining glucose values from the blood of a subject;

(b) comparing the glucose values to a plurality of defined glycemia zones defined as a function of blood glucose concentration; and

(c) automatically delivering insulin or insulin analog via an insulin pump comprising a controller having a multi-zone-model predictive control algorithm comprising the glycemia zones.

18. The method of claim 16 wherein the algorithm provides:

(a) a relatively conservative control action for a large deviation from a normal or desired zone, and

(b) a relative aggressive control action for within the normal or desired zone.

19. The method of claim 16 wherein the physiological zones are blood glucose concentration zones, the drug is insulin, and the algorithm provides:

(a) a relative conservative control action for a elevated “

hyperglycemia”

zone;

(b) a relatively aggressive control action for a normal or near normal “

euglycemia”

zone;

(c) control inaction for a normoglycemia zone; and

(d) a control action that results in suspending insulin delivery for a hypoglycemia zone.

20. The method of claim 16 wherein the algorithm provides glycemia zones and control weights ratios (Q_k and R_k) according to the following schedule: