Seizure onset classification and stimulation parameter selection

• US 10,130,813 B2
• Filed: 02/10/2016
• Issued: 11/20/2018
• Est. Priority Date: 02/10/2015
• Status: Active Grant

• Current Assignee: Neuropace Inc.
• Original Assignee: Neuropace Inc.
• Inventors: Crowder, Tara Leigh, Arcot Desai, Sharanya, Morrell, Martha Jo, Tcheng, Thomas Kim, Kapur, Ritu
• Primary Examiner(s): Fairchild, Mallika D
• Application Number:
  US15/040,932
• Publication Number:
  US 20160228705A1
• Time in Patent Office: 1,014 Days
• CPC Class Codes: A61B 5/374 Detecting the frequency distribution of signals, e.g. detecting delta, theta, alpha, beta or gamma waves
• CPC Class Codes: A61B 5/4094 Diagnosing or monitoring seizure diseases, e.g. epilepsy
• CPC Class Codes: A61B 5/4836 Diagnosis combined with treatment in closed-loop systems or methods A61B5/0036 takes precedence
• CPC Class Codes: A61B 5/7264 Classification of physiological signals or data, e.g. using neural networks, statistical classifiers, expert systems or fuzzy systems neural networks per se G06N3/00; expert systems per se G06N5/00
• CPC Class Codes: A61B 5/7282 Event detection, e.g. detecting unique waveforms indicative of a medical condition A61B5/4094, A61B5/4818, A61B5/349, A61B5/0823 take precedence
• CPC Class Codes: A61N 1/0529 Electrodes for brain stimulation
• CPC Class Codes: A61N 1/36064 Epilepsy
• CPC Class Codes: A61N 1/36139 with automatic adjustment
• CPC Class Codes: A61N 1/37241 providing test stimulations
• CPC Class Codes: A61N 1/37247 User interfaces, e.g. input or presentation means

Abstract

A neurostimulation system senses electrographic signals from the brain of a patient, extracts features from the electrographic signals, and when the extracted features satisfy certain criteria, detects a neurological event type. A mapping function relates the detected neurological event type to a stimulation parameter subspace and a default stimulation parameter set where the values of the stimulation parameters define an instance of stimulation therapy for the patient. The decision whether to implement a stimulation parameter subspace or a default stimulation parameter set may be informed by integrating other information about a state of the patient. A stimulation parameter subspace or stimulation parameter set may optimized by testing it against various thresholds until certain effectiveness criteria is satisfied. The neurological event type may be one of several electrographic seizure onset types.

Claims

1. An implantable neurostimulation system configured to optimize values of parameters which define electrical stimulation delivered to a brain of a patient as a form of neural modulation therapy based on a classification of a sensed brain signal into a neurological event type, the implantable neurostimulation system comprising:

at least one lead provided with, at or near a distal end thereof, a plurality of sensors and a plurality of stimulators, the at least one lead configured to be implanted at least partially in the patient so that the plurality of sensors and the plurality of stimulators are in or on the brain;

an implantable neurostimulator coupled to the at least one lead and configured to, with each of the plurality of sensors, selectively sense neurological activity from the brain and further configured to, with each of the plurality of stimulators, selectively stimulate the brain, the implantable neurostimulator comprising;

a sensing module configured to;

process information received on each of a plurality of sensing channels associated with an input from at least one of the plurality of sensors;

create a record of electrographic activity for each sensing channel;

a feature extraction module configured to extract a plurality of features from each record of electrographic activity;

a mapping function module configured to attempt to match each record of electrographic activity, based on the plurality of extracted features, to one or more of a plurality of neurological event types, wherein each of the plurality of neurological event types is associated with a different parameter configuration comprising;

a stimulation parameter subspace comprising a range of potential values for each of a plurality of stimulation parameters; and

a discrete default value for each of the plurality of stimulation parameters;

wherein the values of the stimulation parameters collectively define an instance of stimulation;

a control module configured to determine which value to set for each of the plurality of stimulation parameters; and

a stimulation module configured to generate the instance of stimulation and to deliver, through a selected one or more of the plurality of stimulators, a plurality of instances of stimulation comprising a course of stimulation therapy, to the patient.

2. The implantable neurostimulator of claim 1, further comprising a memory, wherein the sensing module is further configured to store in the memory, at least temporarily, a file comprising a set of records of electrographic activity for the plurality of sensing channels.

3. The implantable neurostimulation system of claim 1 further comprising:

a communications module configured to wirelessly receive instructions and other information from and to transmit information to an external component.

4. The implantable neurostimulation system of claim 1, wherein the mapping function module comprises a lookup table stored in a memory in which each of the plurality of neurological event types is correlated to a different stimulation parameter subspace.

5. The implantable neurostimulation system of claim 1, wherein the plurality of neurological event types comprises one or more of:

a low-voltage-fast seizure onset, a high-voltage-beta seizure onset, a hypersynchronous seizure onset, a spike-and-wave seizure onset, a multiple seizure onset, an attenuation seizure onset, and a neurological event type identified by a clustering process.

6. The implantable neurostimulation system of claim 1, wherein the stimulation parameter subspace comprises one of:

a range of values comprising high frequencies for a frequency stimulation parameter, a range of values comprising very high frequencies for a frequency stimulation parameter;

a theta burst stimulation parameter subspace, a voltage-controlled, low frequency stimulation parameter subspace, a high amplitude stimulation parameter subspace, and an alerting stimulation parameter subspace.

7. The implantable neurostimulation system of claim 1, wherein the implantable neurostimulator further comprises:

an integration module configured to;

analyze at least one data element other than a record of electrographic activity, which data element corresponds to one or more states of the patient in order to determine a measure of effectiveness of a course of stimulation therapy; and

communicate the measure of effectiveness to the control module,

wherein the control module is further configured to determine which value to set for each of the plurality of stimulation parameters based in part on the measure of effectiveness.

8. The implantable neurostimulation system of claim 7, wherein the at least one data element other than a record of electrographic activity comprises one or both of an electrophysiological metric and a non-electrophysiological metric.

9. The implantable neurostimulation system of claim 8, wherein the electrophysiological metric comprises one or more of:

a detected event count, a detected event rate, a long episode rate, a detected event duration, a saturation rate, changes in power in specific frequency bands, and amplitude of evoked potentials.

10. The implantable neurostimulation system of claim 8, wherein the non-electrophysiological metric comprises one or more of:

oxygenation concentration, pH/alkaline shift, neurotransmitter levels.

11. The implantable neurostimulation system of claim 7, wherein the implantable neurostimulator further comprises:

an optimization module configured to;

receive the measure of effectiveness from the integration module;

compare the measure of effectiveness to a criterion; and

when the criterion has not been satisfied, send an instruction to the control module to set a new value for one or more of the plurality of stimulation parameters.

12. A method of optimizing values of parameters which define electrical stimulation delivered to a patient as a neural modulation therapy by a neurostimulation system, the method comprising:

analyzing, with an analyzer module of an implantable neurostimulator of the neurostimulation system, features extracted from a record of electrographic activity sensed from a plurality of sensors located in or on a patient'"'"'s brain;

mapping, with a mapping function module of either the implantable neurostimulator or an external component of the neurostimulation system, the features extracted from the record of electrographic activity to one of a plurality of different neurological event types, by calculating a probability that the features extracted from the record of electrographic activity correspond most closely to the features of each neurological event type;

wherein each of the plurality of different neurological event types is associated with a different parameter configuration comprising;

a stimulation parameter subspace comprising a range of potential values for each of a plurality of stimulation parameters; and

a discrete default value for each of the plurality of stimulation parameters;

wherein the discrete values of the stimulation parameters collectively define a stimulation parameter set for generating and delivering an instance of stimulation,

determining, with a control module of either the implantable neurostimulator or an external component of the neurostimulation system, a stimulation parameter set based at least in part on the mapping;

generating, with a stimulation module of the implantable neurostimulator, an instance of stimulation according to the determined stimulation parameter set; and

delivering, through a selected one or more a plurality of stimulators, an instance of stimulation to the patient corresponding to the determined stimulation parameter set comprising a stimulation therapy.

13. The method of claim 12, further comprising:

analyzing, with an integration module of either the implantable neurostimulator or an external component of the neurostimulation system, at least one data element other than a record of electrographic activity, which data element corresponds to one or more states of the patient in order to determine a measure of effectiveness of a stimulation therapy;

communicating the measure of effectiveness to the control module; and

determining, with the control module, a stimulation parameter based in part on the measure of effectiveness.

14. The method of claim 13, further comprising:

an optimization module configured to;

receive, in an optimization module of either the implantable neurostimulator or an external component of the neurostimulation system, the measure of effectiveness;

compare, in the optimization module, the measure of effectiveness to a criterion; and

when the criterion has not been satisfied, send an instruction to the control module to set a new value for one or more of the plurality of stimulation parameters.

15. A neurostimulation system configurable to optimize values of parameters which define electrical stimulation delivered to a brain of a patient based on a classification of a sensed brain signal into a neurological event type, the neurostimulation system comprising:

an implantable neurostimulator comprising;

a plurality of sensors and a plurality of stimulators configured to be implanted at least partially in or on the patient'"'"'s brain;

a sensing module coupled to the plurality of sensors and configured to sense electrographic activity from the brain through the plurality of sensors, to process information corresponding to the electrographic activity received on each of a plurality of sensing channels associated with at least one of the plurality of sensors, and to create a record of electrographic activity for each sensing channel;

a feature extraction module configured to extract a plurality of features from each record of electrographic activity;

a mapping function module configured to attempt to match each record of electrographic activity, based on the plurality of extracted features, to one or more of a plurality of neurological event types, wherein each of the plurality of neurological event types is associated with a different parameter configuration comprising;

a stimulation parameter subspace comprising a range of potential values for each of a plurality of stimulation parameters; and

a discrete default value for each of the plurality of stimulation parameters;

wherein the values of the stimulation parameters collectively define an instance of stimulation

a control module configured to determine which value to set for each of the plurality of stimulation parameters;

a stimulation module selectively coupled to the plurality of stimulators and configured to generate the instance of stimulation and to deliver the instance of stimulation, through one or more of the plurality of stimulators; and

a communications module;

an external component configured to provide instructions and other information to, and to receive information transmitted from the communications module of the implantable neurostimulator;

a database for storing information transmitted from the communications module of the implantable neurostimulator; and

a database interface configured to allow a user to view, interact with, and modify information in the database.

16. A method of optimizing values of parameters which define electrical stimulation delivered to a brain of a patient based on a classification of a sensed brain signal into a neurological event type, the method comprising:

sensing electrographic activity from the brain;

processing information corresponding to the sensed electrographic activity to create a record of electrographic activity;

extracting a plurality of features from the record of electrographic activity;

matching the record of electrographic activity, based on the plurality of extracted features, to one or more of a plurality of neurological event types, wherein each of the plurality of neurological event types is mapped to a different parameter configuration comprising;

a stimulation parameter subspace comprising a range of potential values for each of a plurality of stimulation parameters; and

a discrete default value for each of the plurality of stimulation parameters;

determining which value to set for each of the plurality of stimulation parameters;

generating an instance of stimulation based on the set values of each of the plurality of stimulation parameters; and

delivering the instance of stimulation, through one or more of a plurality of stimulators.