Blended posture state classification and therapy delivery

• US 9,327,129 B2
• Filed: 04/30/2009
• Issued: 05/03/2016
• Est. Priority Date: 07/11/2008
• Status: Active Grant

• Current Assignee: Medtronic Incorporated (Medtronic Plc)
• Original Assignee: Medtronic Incorporated (Medtronic Plc)
• Inventors: Panken, Eric J.
• Primary Examiner(s): Stigell, Theodore
• Assistant Examiner(s): Mahmood, Nadia A
• Application Number:
  US12/433,038
• Publication Number:
  US 20100010382A1
• Time in Patent Office: 2,560 Days
• Field of Search: 607 17- 19, 607/62, 607/64
• US Class Current: 1/1
• CPC Class Codes: A61B 2562/0219 Inertial sensors, e.g. accelerometers, gyroscopes, tilt switches
• CPC Class Codes: A61B 5/1116 Determining posture transitions
• CPC Class Codes: A61B 5/1118 Determining activity level
• CPC Class Codes: A61B 5/686 Permanently implanted devices, e.g. pacemakers, other stimulators, biochips A61B5/6861 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 1/36062 Spinal stimulation
• CPC Class Codes: A61N 1/36139 with automatic adjustment
• CPC Class Codes: A61N 1/36535 controlled by body position or posture
• CPC Class Codes: A61N 1/36542 controlled by body motion, e.g. acceleration
• CPC Class Codes: A61N 1/37247 User interfaces, e.g. input or presentation means
• CPC Class Codes: A61N 1/3925 Monitoring; Protecting
• CPC Class Codes: G16H 50/20 for computer-aided diagnosis, e.g. based on medical expert systems

Abstract

Techniques for classification of a posture state of a patient using multiple posture state definitions, and for delivering therapy according to the patient'"'"'s classified posture state. Detected parameter values describing a patient'"'"'s posture state are compared to posture state definitions. This comparison is used to determine similarity values describing how similar the patient'"'"'s posture state is to each of the posture states described by the posture state definitions. Weighting factors may be determined from the similarity values and used to weight therapy parameter values that are associated with each of the posture state definitions. The resulting weighted therapy parameter values may be used to derive a blended therapy parameter value for use in delivering therapy to the patient. The patient'"'"'s posture state may be expressed in terms of a blending of the multiple posture state definitions.

Claims

1. A method, comprising:

obtaining a detected parameter value for use in classifying a posture state of a patient;

for each of multiple posture state definitions, obtaining a respective defined parameter value and a respective therapy parameter value;

determining, for each of a plurality of the posture state definitions, a similarity between the detected parameter value and the respective defined parameter value;

determining a blended therapy parameter value based on the similarities and respective therapy parameter values for the plurality of the posture state definitions; and

delivering therapy to the patient via a medical device according to the blended therapy parameter value.

2. The method of claim 1, wherein the detected parameter value is indicative of a posture of the patient, each of the defined parameter values is indicative of a respective defined posture, and wherein determining, for each of a plurality of the posture state definitions, a similarity comprises determining, for each of the plurality of the posture state definitions, how close the posture of the patient is to the defined posture that is indicated by the respective defined parameter value.

3. The method of claim 2, wherein the detected parameter value comprises a detected posture vector, wherein the respective defined parameter value comprises a respective defined posture vector, and wherein determining, for each of a plurality of the posture state definitions, a similarity further comprises determining, for each of the plurality of the posture state definitions, a cosine of an angle between the detected posture vector and the respective defined posture vector.

4. The method of claim 1, wherein determining, for each of a plurality of the posture state definitions, a similarity between the detected parameter value and the respective defined parameter value comprises selecting the plurality of the posture state definitions.

5. The method of claim 1, further comprising determining a respective weighting factor W for each of the similarities and wherein determining a blended therapy parameter value comprises using the respective weighting factor to weight a respective therapy parameter value.

6. The method of claim 5, wherein each weighting factor is determined using a weighting function.

7. The method of claim 6, wherein determining, for each of a plurality of the posture state definitions, a similarity between the detected parameter value and the respective defined parameter value comprises determining a distance between the detected parameter value and the respective defined parameter value, and further comprising selecting the weighting function to map the distance to a value between 0 and a predetermined maximum positive number.

8. The method of claim 5, wherein each of the plurality of posture state definitions is included in a set I and the respective therapy parameter value is a therapy parameter value P, and wherein the blended therapy parameter value is determined as

9. The method of claim 1, further comprising selecting a subset of multiple positive ones of the similarities for use in determining the blended therapy parameter value.

10. The method of claim 1, further comprising:

determining a weighting factor for each of the similarities;

normalizing each of the weighting factors; and

classifying the posture state based on the normalized weighting factors.

11. The method of claim 1, further comprising:

using selected ones of the similarities to determine time-averaged weighting factors;

normalizing all of the time-averaged weighting factors with respect to each other; and

determining from any one of the normalized time-averaged weighting factors an amount of time attributable to an associated one of the multiple posture state definitions over a predetermined period of time.

12. The method of claim 1, wherein the detected parameter value is indicative of a posture of the patient, and further comprising:

obtaining a second detected parameter value for use in classifying an activity state of a patient;

determining, for each of a second plurality of the posture state definitions, a similarity between the second detected parameter value and the respective defined parameter value;

using the similarity between the second detected parameter value and the respective defined parameter value to weight the respective therapy parameter value;

determining a second blended therapy parameter value based the weighted therapy parameter values; and

delivering therapy to the patient according to the blended therapy parameter value and the second blended therapy parameter value.

13. The method of claim 1, further comprising delivering electrical stimulation therapy to the patient by an implantable medical device according to the blended therapy parameter value.

14. The method of claim 1, wherein the detected parameter value is indicative of motion of the patient, and wherein the blended therapy parameter value is adapted to deliver therapy to the patient in response to the motion.

15. A system, comprising:

a sensor configured to provide a detected parameter value indicative of a posture state of a patient;

a storage device configured to store posture state definitions, each posture state definition being associated with a defined parameter value and a therapy parameter value; and

a processor configured to perform the steps of;

determining a similarity between the detected parameter value and each of a plurality of the defined parameter values associated with a plurality of the posture state definitions; and

determining a blended therapy parameter value for use in delivering therapy to the patient based on the similarities and on the therapy parameter values associated with the plurality of the posture state definitions.

16. The system of claim 15, wherein the detected parameter value indicates a detected posture vector, the defined parameter values each indicates a defined posture vector, and wherein the processor is further configured to determine the similarity based on how close the detected posture vector is to a defined posture vector.

17. The system of claim 16, wherein the processor is further configured to determine the similarity based on a cosine of an angle between the detected posture vector and a defined posture vector.

18. The system of claim 15, wherein the processor is further configured to perform the steps of:

determining a respective weighting factor from each similarity;

multiplying each weighting factor by a therapy parameter value for an associated posture state definition to obtain a weighted therapy parameter value;

obtaining a sum of all of the weighted therapy parameter values; and

determining the blended therapy parameter value by dividing the obtained sum by a sum of all of the weighting factors.

19. The system of claim 15, further comprising a therapy module configured to deliver therapy to the patient based on the blended therapy parameter value.

20. The system of claim 15, wherein the storage device is configured to store weighting functions and wherein the processor is further configured to perform the steps of:

selecting one of the weighting functions;

employing the selected weighting function to determine for each of selected ones of the similarities a respective weighting factor; and

determining the blended therapy parameter value based on the weighting factors.

21. The system of claim 15, wherein the processor is further configured to perform the steps of:

determining for a similarity a respective weighting factor;

time-averaging the weighting factor; and

employing the time-averaged weighting factor to obtain an amount of time attributable to an associated defined posture state.

22. The system of claim 15, wherein the posture state definitions include at least one of activity state definitions and posture definitions.

23. A non-transitory storage medium storing executable instructions to cause a processing circuit to perform the steps, comprising:

receiving a detected parameter value indicative of a posture of a patient;

obtaining multiple posture definitions describing defined postures;

obtaining, for each of the multiple posture definitions, a respective therapy parameter value;

determining multiple similarities, each describing how similar the detected parameter value is to a defined parameter value of a respective posture definition;

obtaining for each of the posture definitions a weighted therapy parameter value that is based on the similarity determined for the posture definition and the respective therapy parameter value obtained for the posture definition; and

obtaining a blended therapy parameter value based on the weighted therapy parameter values.

24. The non-transitory storage medium of claim 23, wherein the steps further comprise controlling delivery of therapy according to at least one of the blended therapy parameter value and a therapy parameter value obtained for a respective one of the postures definitions.

25. The non-transitory storage medium of claim 23, wherein the steps further comprise:

receiving a detected parameter value indicative of an activity level of a patient; and

modifying the blended therapy parameter value to take into account the activity level of the patient.

26. The non-transitory storage medium of claim 23, wherein the steps further comprise:

classifying the posture of the patient based on the posture definitions; and

controlling delivery of therapy to the patient based on the blended therapy parameter value if the patient is not classified as being in any of the defined postures, and otherwise, controlling delivery of therapy to the patient based on a therapy parameter value obtained for a posture definition describing the posture in which the patient is classified.

27. The non-transitory storage medium of claim 23, wherein the steps further comprise:

selecting a weighting function;

using the weighting function to determine a respective weighting factor for each of the similarities; and

for each of the posture definitions, obtaining a weighted therapy parameter value for the posture definition by multiplying the respective weighting factor by the respective therapy parameter value obtained for the posture definition.

28. A method, comprising:

obtaining from a sensor carried by a patient a signal indicative of a posture of the patient;

comparing the signal to each of multiple posture definitions to obtain multiple similarities, each describing how similar the posture is to a respective posture definition;

providing a classification of the posture that is based on the multiple similarities; and

providing care to the patient via a medical device based on the classification.

29. The method of claim 28, wherein providing care to the patient comprises providing therapy to the patient via an implantable medical device based on the classification.

30. The method of claim 28, further comprising storing the classification for analyzing a condition of the patient.

31. The method of claim 28, wherein the classification of the postures is based on an array of values, each of the values being derived from a respective one of the multiple similarities and describing a portion of the posture attributable to a respective posture definition.

32. The method of claim 28, wherein providing care to the patient comprises providing neurostimulation therapy to the patient via an implantable medical device based on the classification.