Systems and methods for spinal cord stimulation

• US 10,213,605 B2
• Filed: 03/09/2017
• Issued: 02/26/2019
• Est. Priority Date: 09/26/2014
• Status: Active Grant

• Current Assignee: Duke University
• Original Assignee: Duke University
• Inventors: Grill, Warren M., Zhang, Tianhe
• Primary Examiner(s): Manuel, George
• Application Number:
  US15/454,484
• Publication Number:
  US 20170182322A1
• Time in Patent Office: 719 Days
• CPC Class Codes: A61N 1/36071 Pain
• CPC Class Codes: A61N 1/36132 using patient feedback
• CPC Class Codes: A61N 1/36139 with automatic adjustment
• CPC Class Codes: A61N 1/37217 characterised by the communication link, e.g. acoustic or tactile
• CPC Class Codes: A61N 1/37247 User interfaces, e.g. input or presentation means
• 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

Optimizing temporal pulse patterns for stimulation delivery to a subject includes: generating a first generation pulse pattern; delivering stimulation according to the first generation pattern to a subject; measuring efficacy, efficiency and side-effect parameters affected by the delivered stimulation; determining a fitness of the first generation pattern using the measured parameters; generating a second generation pattern using the first generation pattern according to the determined fitness of the first generation temporal pattern; and delivering stimulation according to the second generation pattern. Iterative further optimization may include crossing any particular generation temporal pulse pattern with at least one other temporal pulse pattern to generate offspring patterns for further use and optimization. Immigrant random patterns may be added, and offspring patterns may receive point mutations. Optimization may be conducted using a computational model prior to treatment of a human subject and further optimization may occur using feedback from the subject.

Claims

1. A method of optimizing temporal pulse patterns for stimulation delivery to a subject, the method comprising:

generating a first generation temporal pulse pattern;

delivering stimulation according to the first generation temporal pulse pattern to a subject;

measuring efficacy, efficiency and side-effect parameters affected by the delivered stimulation;

determining a fitness of the first generation temporal pulse pattern using the measured efficacy, efficiency and side-effect parameters;

generating a second generation temporal pulse pattern using the first generation temporal pulse pattern according to the determined fitness of the first generation temporal pulse pattern; and

delivering stimulation according to the second generation temporal pulse pattern to the subject,

wherein generating the second generation temporal pulse pattern comprises using a genetic algorithm that uses an output of a computational model including representations of dorsal horn wide-dynamic range (WDR) neurons responsible for transmitting nociceptive information and dorsal column nuclei (DCN) neurons associated with side-effect paresthesias.

2. A method according to claim 1, further comprising iteratively:

delivering stimulation to the subject according to any particular generation temporal pulse pattern,

measuring efficacy, efficiency and side-effect parameters affected by delivery of the stimulation to the subject according to the particular generation temporal pulse pattern;

determining a fitness of the particular generation temporal pulse pattern using the measured efficacy, efficiency and side-effect parameters thereof; and

generating a next generation temporal pulse pattern using the particular generation temporal pulse pattern according to the determined fitness of the particular generation temporal pulse pattern.

3. A method according to claim 2, wherein generating the next generation temporal pulse pattern comprises crossing the particular generation temporal pulse pattern with at least one other temporal pulse pattern.

4. A method according to claim 3, wherein a probability of any pulse in the next generation temporal pulse pattern originating from either the particular generation temporal pulse pattern or the at least one other temporal pulse pattern is independent of that of any other pulse.

5. A method according to claim 3, wherein:

crossing the particular generation temporal pulse pattern with at least one other temporal pulse pattern generates an offspring temporal pulse pattern, and

generating the next generation temporal pulse pattern comprises adding at least one point mutation to the offspring temporal pulse pattern.

6. A method according to claim 2, wherein generating a next generation temporal pulse pattern further comprises using random patterns.

7. A method according to claim 2, wherein:

the subject comprises a physical subject; and

delivering stimulation to the subject comprises delivering electrical pulses to the physical subject.

8. A method according to claim 2, wherein:

the subject comprises the computational model; and

delivering stimulation to the subject comprises simulating delivery of pulses to the computational model.

9. A method according to claim 8, further comprising, for each simulated delivery of pulses to the computational model, measuring efficacy, efficiency and side-effect parameters.

10. A method according to claim 8, wherein the computational model includes a network of simulated biophysical neurons connected in a manner consistent with a physical dorsal horn pain processing neuron network.

11. A method according to claim 10, wherein the computational model incorporates delays based on the conduction velocities of neuron fibers to simulate signal propagation from a peripheral or dorsal column nerve fiber.

12. A method according to claim 1, wherein:

the first generation temporal pulse pattern is generated by simulating delivery of pulses to the computational model; and

delivering stimulation according to the first generation temporal pulse pattern to a subject comprises delivering electrical pulses to a physical subject.

13. A method according to claim 1, wherein measuring efficacy, efficiency and side-effect parameters affected by the delivered stimulation comprises measuring efficacy by determining reduction of neural activity associated with pain relief.

14. A method according to claim 1, wherein measuring efficacy, efficiency and side-effect parameters affected by the delivered stimulation comprises measuring an efficacy parameter by determining the difference between an average firing rate of a neuron while delivering stimulation according to the first generation temporal pulse pattern and the firing rate of the neuron during constant frequency stimulation with which the first generation temporal pulse pattern has the same average frequency.

15. A method according to claim 1, wherein measuring efficacy, efficiency and side-effect parameters affected by the delivered stimulation comprises measuring an efficiency parameter by determining the average frequency of the first generation temporal pulse pattern.

16. A method according to claim 1, wherein measuring efficacy, efficiency and side-effect parameters affected by the delivered stimulation comprises measuring a side effect parameter by determining an average side effect firing rate of neurons.

17. A method according to claim 1, wherein the subject is a human patient and the side effect parameter is determined at least in part using feedback from the human patient.

18. A method according to claim 1, wherein generating a first generation temporal pulse pattern comprises iteratively generating multiple prior generations of temporal pulse patterns using a computational model.

19. A method according to claim 18, wherein, each prior generation temporal pulse pattern is used to deliver simulated stimulation to the computational model and efficacy, efficiency and side-effect parameters are measured by calculation to determine fitness for use of the prior generation temporal pulse pattern in generating subsequent generation temporal pulse patterns.

20. A method according to claim 1, wherein each temporal pulse pattern comprises pulses each having an adjustable duration and amplitude.

21. A method according to claim 1, wherein:

generating a first generation temporal pulse pattern and a second generation temporal pulse pattern comprises using a computational model by simulating delivery of pulse patterns to the computational model and iteratively constructing successive generations of the pulse patterns by determining fitness of each particular pulse pattern and crossing the pulse patterns that are determined to be fit; and

,

wherein, the method further includes saving the generated temporal pulse pattern sets for selection by a user; and

wherein, delivering stimulation to the subject is made according to a selection made by the user of one or more of the saved temporal pulse pattern sets.