Electrode selection for electrical stimulation therapy

Patent metadata

Field	Value
Publication number	US-2016144186-A1
Application number	US-201514724327-A
Country	US
Kind code	A1
Filing date	May 28, 2015
Priority date	Nov 21, 2014
Publication date	May 26, 2016
Grant date	—

Abstract

In one example, a method includes selecting, by one or more processors and based on a representation of sensed electrical signals for a particular patient and a plurality of representations of sensed electrical signals for a plurality of other patients, a combination of electrodes of a plurality of combinations of one or more implantable electrodes for delivery of electrical stimulation to the particular patient.

First claim

What is claimed is:

1 . A method for controlling delivery of electrical stimulation therapy, the method comprising:
selecting, by one or more processors and based on a representation of sensed electrical signals for a particular patient and a plurality of representations of sensed electrical signals for a plurality of other patients, a combination of electrodes of a plurality of combinations of one or more implantable electrodes for delivery of electrical stimulation therapy to the particular patient; and generating, by the one or more processors, information to control the delivery of the electrical stimulation based on the selected combination of the electrodes.

2 . The method of claim 1 , wherein each respective representation of sensed electrical signals of the plurality of representations of sensed electrical signals is associated with a respective combination of electrodes of the plurality of electrodes selected for delivery of electrical stimulation for each respective patient of the plurality of patients.

3 . The method of claim 2 , wherein selecting the combination of electrodes comprises:
selecting, by the one or more processors and using one or more machine learning models trained based on one or more of the plurality of representations of sensed electrical signals, the combination of electrodes for delivery of electrical stimulation to the particular patient.

4 . The method of claim 3 , wherein the one or more machine learning models comprise a plurality of support vector machines (SVMs) that are each associated with a particular combination of electrodes of the plurality of electrodes, and wherein selecting the combination of electrodes comprises:
determining, for each respective SVM of the plurality of SVMs, a respective score that indicates a degree to which the representation of sensed electrical signals for the particular patient resembles that of the sensed electrical signals from a plurality of patients for whom the respective combination of electrodes that is associated with the respective SVM has been therapeutically effective for delivery of electrical stimulation; and selecting, based on the scores, the combination of electrodes to deliver electrical stimulation to the particular patient.

5 . The method of claim 3 , wherein the machine learning models are trained based on signals in a theta band and signals in a beta band of the plurality of representations of electrical signals, and wherein selecting the combination of electrodes comprises:
selecting, based on the signals in the theta band and the signals in the beta band of the representation of electrical signals for the particular patient, the combination of electrodes to deliver electrical stimulation to the particular patient.

6 . The method of claim 3 , wherein the machine learning models are trained based on signals in a theta band of the plurality of representations of electrical signals, and wherein selecting the combination of electrodes comprises:
selecting, based on the signals in the theta band of the representation of electrical signals for the particular patient, the combination of electrodes to deliver electrical stimulation to the particular patient.

7 . The method of claim 3 , wherein the machine learning models are trained based on signals in a beta band of the plurality of representations of electrical signals, and wherein selecting the combination of electrodes comprises:
selecting, based on the signals in the beta band of the representation of electrical signals for the particular patient, the combination of electrodes to deliver electrical stimulation to the particular patient.

8 . The method of claim 5 , wherein the machine learning models are trained based on fewer representations of electrical signals than all the possible representations of electrical signals from all possible combinations of electrodes of the plurality of combinations of electrodes.

9 . The method of claim 3 , wherein the one or more machine learning models are further trained based on one or more representations of sensed electrical signals for the particular patient and corresponding combinations of electrodes that were therapeutically effective for the particular patient.

10 . The method of claim 1 , wherein the one or more processors are included in a device implanted in the particular patient configured to deliver electrical stimulation to the particular patient using the one or more implantable electrodes, the method further comprising:
determining, by the one or more processors, the representation of electrical signals for the particular patient based on electrical signals measured across one or more combinations of contacts of the plurality of combinations of electrodes.

11 . The method of claim 10 , wherein determining the representation of electrical signals for the particular patient comprises periodically determining an updated representation of electrical signals for the particular patient, the method further comprising:
automatically selecting, by the one or more processors and based on the updated representation of electrical signals for the particular patient, an updated combination of electrodes of the plurality of combinations of electrodes for delivery of electrical stimulation; and delivering, by the device, electrical stimulation to the particular patient via the updated combination of electrodes.

12 . The method of claim 1 , wherein the one or more processors are included in an external programmer of a device implanted in the particular patient, wherein generating the information to control the delivery of the electrical stimulation based on the selected combination of the electrodes comprises:
programming, by the external programmer, the device implanted in the particular patient to deliver the electrical stimulation to the particular patient via the selected combination of electrodes.

13 . A device comprising:
a memory configured to store a representation of sensed electrical signals for a particular patient; and one or more processors configured to:
select, based on the representation of sensed electrical signals for a particular patient and a plurality of representations of sensed electrical signals for a plurality of other patients, a combination of electrodes of a plurality of combinations of one or more implantable electrodes for delivery of electrical stimulation therapy to the particular patient; and
generate information to control the delivery of the electrical stimulation based on the selected combination of the electrodes.

14 . The device of claim 13 , wherein each respective representation of sensed electrical signals of the plurality of representations of sensed electrical signals is associated with a respective combination of electrodes of the plurality of electrodes selected for delivery of electrical stimulation for each respective patient of the plurality of patients.

15 . The device of claim 14 , wherein, to select the combination of electrodes, the one or more processors are configured to:
select, using one or more machine learning models trained based on one or more of the plurality of representations of sensed electrical signals, the combination of electrodes for delivery of electrical stimulation to the particular patient.

16 . The device of claim 15 , wherein the one or more machine learning models comprise a plurality of support vector machines (SVMs) that are each associated with a particular combination of electrodes of the plurality of electrodes, and wherein, to select the combination of electrodes, the one or more processors are co

Assignees

• Medtronic Inc

Inventors

• Kaemmerer William F
• Dani Siddharth
• Connolly Allison T

Classifications

• A61N1/36064

  Epilepsy · CPC title

• A61N1/36075

  Headache or migraine · CPC title

• A61N1/36067

  Movement disorders, e.g. tremor or Parkinson disease (stimulating motor muscle A61N1/36003) · CPC title

• A61N1/36096

  Mood disorders, e.g. depression, anxiety or panic disorder · CPC title

• A61N1/0534

  Electrodes for deep brain stimulation · CPC title