# US 2026/0241174 A1 · Feature selection and sensing optimization for adaptive neuromodulation therapy

> Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt

- Number: US 2026/0241174 A1
- Document ID: US20260241174A1
- Status: Published application
- Kind: A1
- Application number: 19/641,772
- Filed: 2026-04-08
- Published: 2026-08-20
- Earliest claimed: 2022-08-16
- Assignees: Boston Scientific Neuromodulation Corporation
- Applicants: Boston Scientific Neuromodulation Corporation
- Inventors: Andrew James Haddock; Tianhe Zhang; Satya Venkata Sandeep Avvaru
- Web page: https://patentgrep.com/patent/US20260241174A1

## CPC

- A61N 1/36114: {Cardiac control, e.g. by vagal stimulation (stimulating the heart A61N1/362)}
- A61N 1/36139: {with automatic adjustment}
- A61N 1/36117: {for treating hypertension}
- A61N 1/0551: {Spinal or peripheral nerve electrodes}
- A61N: ELECTROTHERAPY; MAGNETOTHERAPY; RADIATION THERAPY; ULTRASOUND THERAPY (measurement of bioelectric currents A61B; surgical instruments, devices or methods for transferring non-mechanical forms of energy to or from the body A61B18/00; anaesthetic apparatus in general A61M; incandescent lamps H01K; infrared radiators for heating H05B)

## Abstract

Systems and methods for closed-loop control of electrostimulation based on signal features derived from evoked neural activities are disclosed. A system comprises an electrostimulator to deliver a stimulation therapy, a sensing circuit to sense an evoked response induced by the stimulation therapy, and a controller circuit to identify, from a plurality of candidate signal features, a target signal feature that satisfies a performance criterion for distinguishing a first stimulation effect according to a first stimulation setting from a second stimulation effect according to a different second stimulation setting. The target signal feature can be identified based on evaluations of a discrimination metric for each of the candidate signal features. The controller circuit can determine a value of the target signal feature from the sensed evoked response, and control the electrostimulator to adjust the stimulation therapy based on the value of the target signal feature.

## Claims

### Claim 1

A system for providing electrostimulation to a patient, the system comprising:
an electrostimulator configured to generate and deliver a stimulation therapy to a neural target of the patient;
a sensor circuit configured to sense an evoked response induced by the stimulatio2n therapy delivered to the neural target; and
a controller circuit configured to:
identify a target sensing configuration from a plurality of candidate sensing configurations for use with a signal feature, the target sensing configuration satisfying a performance criterion for distinguishing (i) a first stimulation effect according to a first stimulation setting from (ii) a second stimulation effect according to a different second stimulation setting;
determine, from the sensed evoked response obtained in accordance with the target sensing configuration, a value of the signal feature; and
generate a control signal to the electrostimulator to adjust the stimulation therapy based on the value of the signal feature,
wherein, to identify the target sensing configuration that satisfies the performance criterion, the controller circuit is configured to:
for each of the plurality of candidate sensing configurations, determine a set of values of the signal feature from a set of evoked response signals sensed under the candidate sensing configuration, and evaluate a discrimination metric using the set of values of the signal feature, the discrimination metric measuring a performance of the signal feature in distinguishing the first stimulation effect according to the first stimulation setting from the second stimulation effect according to the different second stimulation setting; and
identify, from the plurality of candidate sensing configurations, the target sensing configuration with a corresponding evaluated discrimination metric satisfying a specific condition.

### Claim 2 (depends on 1)

The system of claim 1, wherein the signal feature includes one or more of:
a signal amplitude range within a specified time window;
a signal curve length representing accumulated signal amplitude differences over consecutive unit times within the specified time window; or
a signal power within the specified time window.

### Claim 3 (depends on 1)

The system of claim 1, wherein the first stimulation effect includes a first statistical distribution of the set of values of the signal feature in response to the stimulation therapy according to the first stimulation setting, and the second stimulation effect includes a second statistical distribution of the set of values of the signal feature in response to the stimulation therapy according to the second stimulation setting,
wherein, to evaluate the discrimination metric for each of the plurality of candidate sensing configurations, the controller circuit is configured to determine a degree of overlap between the first statistical distribution and the second statistical distribution.

### Claim 4 (depends on 3)

The system of claim 3, wherein the discrimination metric includes a Fischer's Linear Discriminant (FLD) between the first statistical distribution and the second statistical distribution,
wherein the controller circuit is configured to identify the target sensing configuration with a corresponding FLD value exceeding a threshold or being greater than FLD values of other of the plurality of candidate sensing configurations.

### Claim 5 (depends on 3)

The system of claim 3, wherein the discrimination metric includes a Jensen Shannon Divergence (JSD) metric between the first statistical distribution and the second statistical distribution,
wherein the controller circuit is configured to identify the target sensing configuration with a corresponding JSD value exceeding a threshold or being greater than JSD values of other of the plurality of candidate sensing configurations.

### Claim 6 (depends on 1)

The system of claim 1, wherein the plurality of candidate sensing configurations include different sensing electrode configurations, different sensing locations with respect to the neural target, or both.

### Claim 7 (depends on 1)

The system of claim 1, wherein the controller circuit is further configured to perform a quality check on the sensed evoked response before determining the value of the signal feature.

### Claim 8

A system for providing electrostimulation to a patient, the system comprising:
an electrostimulator configured to generate and deliver a stimulation therapy to a neural target of the patient;
a sensor circuit configured to sense an evoked response induced by the stimulation therapy delivered to the neural target; and
a controller circuit configured to:
identify a target signal pre-processing setting from a plurality of candidate signal pre-processing settings for use with a signal feature, the target signal pre-processing setting satisfying a performance criterion for distinguishing (i) a first stimulation effect according to a first stimulation setting from (ii) a second stimulation effect according to a different second stimulation setting;
determine, from the sensed evoked response pre-processed using the target signal pre-processing setting, a value of the signal feature; and
generate a control signal to the electrostimulator to adjust the stimulation therapy based on the value of the signal feature,
wherein, to identify the target signal pre-processing setting that satisfies the performance criterion, the controller circuit is configured to:
for each of the plurality of candidate signal pre-processing settings, determine a set of values of the signal feature from a set of evoked response signals pre-processed using the candidate signal pre-processing setting, and evaluate a discrimination metric using the set of values of the signal feature, the discrimination metric measuring a performance of the signal feature in distinguishing the first stimulation effect according to the first stimulation setting from the second stimulation effect according to the different second stimulation setting; and
identify, from the plurality of candidate signal pre-processing settings, the target signal pre-processing setting with a corresponding evaluated discrimination metric satisfying a specific condition.

### Claim 9 (depends on 8)

The system of claim 8, wherein the plurality of candidate signal pre-processing settings include:
a plurality of signal filtering settings;
a plurality of epoch averaging settings;
a plurality of artifact treatment settings; or
a plurality of feature window settings.

### Claim 10 (depends on 8)

The system of claim 8, wherein the first stimulation effect includes a first statistical distribution of the set of values of the signal feature in response to the stimulation therapy according to the first stimulation setting, and the second stimulation effect includes a second statistical distribution of the set of values of the signal feature in response to the stimulation therapy according to the second stimulation setting,
wherein, to evaluate the discrimination metric for each of the plurality of candidate signal pre-processing settings, the controller circuit is configured to determine a degree of overlap between the first statistical distribution and the second statistical distribution.

### Claim 11 (depends on 10)

The system of claim 10, wherein the discrimination metric includes a Fischer's Linear Discriminant (FLD) between the first statistical distribution and the second statistical distribution,
wherein the controller circuit is configured to identify the target signal pre-processing setting with a corresponding FLD value exceeding a threshold or being greater than FLD values of other of the plurality of candidate signal pre-processing settings.

### Claim 12 (depends on 10)

The system of claim 10, wherein the discrimination metric includes a Jensen Shannon Divergence (JSD) metric between the first statistical distribution and the second statistical distribution,
wherein the controller circuit is configured to identify the target signal pre-processing setting with a corresponding JSD value exceeding a threshold or being greater than JSD values of other of the plurality of candidate signal pre-processing settings.

### Claim 13 (depends on 8)

The system of claim 8, wherein the signal feature includes one or more of:
a signal amplitude range within a specified time window;
a signal curve length representing accumulated signal amplitude differences over consecutive unit times within the specified time window; or
a signal power within the specified time window.

### Claim 14 (depends on 8)

The system of claim 8, wherein the controller circuit is configured to pre-process the sensed evoked response using the target signal pre-processing setting after the sensed evoked response passes a quality check.

### Claim 15

A method for providing electrostimulation to a patient, the method comprising:
sensing, via a sensor circuit, an evoked response induced by a stimulation therapy delivered to a neural target of the patient using an electrostimulator;
identifying, via a controller circuit, a target configuration from a plurality of candidate configurations, the target configuration comprising a target sensing configuration, a target signal pre-processing setting, or both, the target configuration satisfying a performance criterion for distinguishing (i) a first stimulation effect according to a first stimulation setting from (ii) a second stimulation effect according to a different second stimulation setting;
determining, from the sensed evoked response using the target configuration, a value of a signal feature; and
generating, via the controller circuit, a control signal to the electrostimulator to adjust the stimulation therapy based on the value of the signal feature,
wherein identifying the target configuration includes:
for each of the plurality of candidate configurations, determining a set of values of the signal feature from a set of evoked response signals obtained using the candidate configuration, and evaluating a discrimination metric using the set of values of the signal feature, the discrimination metric measuring a performance of the signal feature in distinguishing the first stimulation effect according to the first stimulation setting from the second stimulation effect according to the different second stimulation setting; and
identifying, from the plurality of candidate configurations, the target configuration with a corresponding evaluated discrimination metric satisfying a specific condition.

### Claim 16 (depends on 15)

The method of claim 15, wherein the target configuration comprises the target sensing configuration, and obtaining the candidate configuration includes sensing respective evoked response signals under respective candidate sensing configurations.

### Claim 17 (depends on 15)

The method of claim 15, wherein the target configuration comprises the target signal pre-processing setting, and obtaining the candidate configuration includes pre-processing respective evoked response signals using respective candidate signal pre-processing settings.

### Claim 18 (depends on 15)

The method of claim 15, wherein the first stimulation effect includes a first statistical distribution of the set of values of the signal feature in response to the stimulation therapy according to the first stimulation setting, and the second stimulation effect includes a second statistical distribution of the set of values of the signal feature in response to the stimulation therapy according to the second stimulation setting,
wherein evaluating the discrimination metric includes determining a degree of overlap between the first statistical distribution and the second statistical distribution.

### Claim 19 (depends on 18)

The method of claim 18, wherein the discrimination metric includes a Fischer's Linear Discriminant (FLD) or a Jensen Shannon Divergence (JSD) between the first statistical distribution and the second statistical distribution,
wherein the target configuration is identified to have a corresponding FLD value or JSD value exceeding a threshold or being greater than FLD values or JSD values of other of the plurality of candidate configurations.

### Claim 20 (depends on 15)

The method of claim 15, further comprising performing a quality check on the sensed evoked response before determining the value of the signal feature using the target configuration.
