Neuromodulation system incorporating multivariate sensing, multivariable pattern recognition, and patient specific adaptation

• US 9,421,379 B2
• Filed: 01/28/2015
• Issued: 08/23/2016
• Est. Priority Date: 02/25/2014
• Status: Active Grant

• Current Assignee: Boston Scientific Neuromodulation Corporation (Boston Scientific Corporation)
• Original Assignee: Boston Scientific Neuromodulation Corporation (Boston Scientific Corporation)
• Inventors: Zhu, Changfang
• Primary Examiner(s): GETZOW, SCOTT M
• Application Number:
  US14/607,188
• Publication Number:
  US 20150238765A1
• Time in Patent Office: 573 Days
• US Class Current: 1/1
• CPC Class Codes: A61B 5/1116 Determining posture transitions
• CPC Class Codes: A61B 5/1118 Determining activity level
• CPC Class Codes: A61B 5/24 Detecting, measuring or recording bioelectric or biomagnetic signals of the body or parts thereof
• CPC Class Codes: A61B 5/407 Evaluating the spinal cord A61B5/4896 takes precedence
• CPC Class Codes: A61B 5/4836 Diagnosis combined with treatment in closed-loop systems or methods A61B5/0036 takes precedence
• CPC Class Codes: A61B 5/6867 specially adapted to be attached or implanted in a specific body part A61B5/6847 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: A61N 1/36071 Pain
• 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 1/37235 Aspects of the external programmer
• CPC Class Codes: G16H 50/20 for computer-aided diagnosis, e.g. based on medical expert systems

Abstract

A medical system configured for performing a medical function in a patient. The medical system comprises at least one medical lead configured for being implanted adjacent a tissue region of the patient, at least one sensing element, monitoring circuitry configured for acquiring a plurality of physiological measurements from the sensing element(s) during one of a plurality of events respectively corresponding to a plurality of classes, and a processor configured for deriving a set of data from the plurality of physiological measurements, performing a feature extraction technique on the data set to acquire at least one feature, analyzing the feature(s), and classifying the data set into the one class corresponding to the one event. The medical system further comprises a controller configured for performing a function based on the classified data set.

Claims

1. A medical system configured for performing a medical function in a patient, comprising:

at least one medical lead configured for being implanted adjacent a tissue region of the patient;

at least one sensing element;

monitoring circuitry configured for acquiring a plurality of physiological measurements from the at least one sensing element during one of a plurality of events respectively corresponding to a plurality of classes;

a processor configured for deriving a set of data from the plurality of physiological measurements, performing a feature extraction technique on the data set to acquire at least one feature, analyzing the at least one feature, and classifying the data set into the one class corresponding to the one event; and

a controller configured for performing a function based on the classified data set.

2. The medical system of claim 1, wherein:

the at least one sensing element comprises a plurality of sensing elements, and the physiological measurements are respectively acquired from the sensing elements;

or

the at least one sensing element is configured for being implanted adjacent the tissue region;

or

the at least one sensing element comprises at least one electrode carried by the at least one medical lead, and the physiological measurements are electrical measurements.

3. The medical system of claim 1, wherein the data set comprises one or both of electrical impedance values and field potential values.

4. The medical system of claim 1, wherein the data set comprises different types of data, wherein the different types of data comprises at least two of an impedance, field potential, evoked potential, pressure, tension, translucence, reflectance, pH, acceleration, chemical, respiration, vascular pulsation, heartbeat, ECG, EKG, and/or EMG, relative lead alignment, absolute lead movement, absolute lead position, and patient posture.

5. The medical system of claim 1, wherein the feature extraction technique is performed using one of a Principal Component Analysis, a Partial Least Square analysis, an Independent Component Analysis, and an artificial neural network.

6. The medical system of claim 1, wherein the at least one feature comprises a plurality of features.

7. The medical system of claim 1, wherein the plurality of events respectively comprises a plurality of different patient postures.

8. The medical system of claim 1, wherein the function is identifying the one event corresponding to the classified data set, and wherein the one event is:

a posture and/or physical activity of the patient;

or

a migration of the at least one medical lead;

or

a stage of an encapsulation process with respect to the at least one medical lead.

9. The medical system of claim 1, wherein the function is a corrective action.

10. The medical system of claim 9, further comprising:

at least one electrode carried by the at least one medical lead; and

analog output circuitry configured for conveying therapeutic electrical energy to the at least one electrode;

wherein the controller is configured for programming the at least one electrode with a set of neurostimulation parameters prior to performing the function, and the corrective action is modifying the neurostimulation parameter set and reprogramming the at least one electrode with the modified neurostimulation parameter set.

11. The medical system of claim 9, wherein the corrective action is providing a notification message or warning to a user.

12. A medical system configured for performing a medical function in a patient, comprising:

at least one medical lead configured for being implanted adjacent a tissue region of the patient;

at least one sensing element;

monitoring circuitry configured for acquiring a plurality of physiological measurements from the at least one sensing element during one of a plurality of events respectively corresponding to a plurality of classes;

a processor configured for deriving a set of data from the plurality of physiological measurements, performing a feature extraction technique on the data set to acquire at least one feature, analyzing the at least one feature, and classifying the data set into the one class corresponding to the one event; and

a controller configured for performing a function based on the classified data set,

wherein each of the at least one feature has a coefficient, the medical system further comprises memory storing at least one (n−

1)-dimensional classification hyperplane that divides an n-dimensional space into a plurality of sub-spaces respectively corresponding to the plurality of different classes, wherein n equals the number of the at least one feature, and wherein the processor is configured for classifying the data set into the one class by recalling the classification hyperplane from the memory, plotting the recalled classification hyperplane and a data point defined by the at least one coefficient in the n-dimensional space, and identifying the sub-space in which the data point is located.

13. A medical system configured for performing a medical function in a patient, comprising:

at least one medical lead configured for being implanted adjacent a tissue region of the patient;

at least one sensing element;

monitoring circuitry configured for acquiring a plurality of physiological measurements from the at least one sensing element during one of a plurality of events respectively corresponding to a plurality of classes;

a processor configured for deriving a set of data from the plurality of physiological measurements, performing a feature extraction technique on the data set to acquire at least one feature, analyzing the at least one feature, and classifying the data set into the one class corresponding to the one event; and

a controller configured for performing a function based on the classified data set,

wherein, for each of the plurality of events, the monitoring circuitry is configured for acquiring at least one plurality of reference physiological measurements from the at least one sensing element, and the processor is configured for deriving at least one set of reference data respectively from the at least one plurality of reference physiological measurements, performing a feature extraction process on each of the at least one set of reference data to acquire a set of reference features for the respective each reference data set, and defining the plurality of classes from the sets of reference features acquired over the plurality of events.

14. The medical system of claim 13, wherein each set of reference features respectively has a set of coefficients, wherein the at least one plurality of reference physiological measurements for each event comprises at least two pluralities of reference physiological measurements, and the processor is configured for defining the plurality of classes by plotting a plurality of data points in an n-dimensional space, wherein n equals the number of the reference features in each set of reference features, each of the data points being defined by the set of coefficients of the set of reference features, wherein the plurality of classes are defined based on the plotted data points.

15. The medical system of claim 14, wherein processor is configured for defining the plurality of classes by performing a pattern recognition technique on the plotted data points.

16. The medical system of claim 15, wherein the pattern recognition technique is performed using one of machine learning, a classification algorithm, a clustering algorithm, a regression algorithm, and an artificial neural network.

17. The medical system of claim 14, wherein the processor is configured for defining at least one (n−

1)-dimensional classification hyperplane that divides the n-dimensional space into a plurality of sub-spaces respectively corresponding to the plurality of classes, the medical system further comprising a memory configured for storing the classification hyperplane, wherein the processor is configured for classifying the data set into the one class by recalling the classification hyperplane from the memory, plotting the recalled classification hyperplane and a data point defined by the at least one coefficient in the n-dimensional space, and identifying the sub-space in which the data point is located.

18. The medical system of claim 13, wherein the processor, for each of the plurality of events, is configured for performing a feature selection technique on a plurality of reference features extracted from each of the at least one set of reference data, and performing a feature selection process that identifies the most significant features that differ between the plurality of events, wherein the acquired set of features includes only the most significant features.

19. The medical system of claim 18, wherein the feature selection technique comprises one of a clustering algorithm, inter-class distance based feature selection, intra-class based feature selection, probabilistic distance based feature selection, correlation based feature selection, consistency based feature selection, and entropy based feature selection.

20. The medical system of claim 13, further comprising:

at least one electrode carried by the at least one medical lead;

telemetry circuitry configured for receiving different user-defined sets of stimulation parameters from an external control device, the different sets of stimulation parameters being respectively correlated to the plurality of events;

analog output circuitry configured for conveying therapeutic electrical energy to the at least one electrode in accordance with the different user-defined sets of stimulation parameters;

wherein the monitoring circuitry is configured for acquiring the at least one plurality of reference physiological measurements from the at least one sensing element in response to the receipt of each of the different user-defined sets of stimulation parameters from the external control device, and wherein the function comprises instructing the analog output circuitry to convey the therapeutic electrical energy to the at least one electrode in accordance with the set of stimulation parameters correlated to the one event.