Methods, systems and apparatuses for detecting seizure and non-seizure states

• US 9,681,836 B2
• Filed: 07/25/2012
• Issued: 06/20/2017
• Est. Priority Date: 04/23/2012
• Status: Active Grant

• Current Assignee: LivaNova USA, Inc. (LivaNova Plc)
• Original Assignee: Cyberonics, Inc
• Inventors: Sabesan, Shivkumar
• Primary Examiner(s): SIMS, JASON M
• Application Number:
  US13/557,351
• Publication Number:
  US 20140031635A1
• Time in Patent Office: 1,791 Days
• Field of Search: None
• CPC Class Codes: A61B 2562/0219 Inertial sensors, e.g. accelerometers, gyroscopes, tilt switches
• CPC Class Codes: A61B 5/0531 Measuring skin impedance
• CPC Class Codes: A61B 5/08 Detecting, measuring or recording devices for evaluating the respiratory organs A61B5/0205 takes precedence
• CPC Class Codes: A61B 5/11 Measuring movement of the entire body or parts thereof, e.g. head or hand tremor, mobility of a limb for measuring pulse A61B5/02 ;A61B5/1038 takes precedence; motion detection to correct for motion artifacts in physiological signals A61B5/721
• CPC Class Codes: A61B 5/318 Heart-related electrical modalities, e.g. electrocardiography [ECG]
• CPC Class Codes: A61B 5/33 specially adapted for cooperation with other devices
• CPC Class Codes: A61B 5/4094 Diagnosing or monitoring seizure diseases, e.g. epilepsy
• CPC Class Codes: A61B 5/7253 characterised by using transforms
• 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

Abstract

Methods, systems, and apparatuses for detecting seizure events are disclosed, having at least one accelerometer to be positioned on a patient and configured to collect acceleration data and a processor in communication with the at least one accelerometer and configured to receive acceleration data from the at least one accelerometer. The processor may apply at least one non-linear operator to the acceleration data to determine whether the acceleration data indicates an event, such application including calculation of a non-linear energy of the acceleration data and performance of at least one secondary analysis to determine whether the event is a seizure.

Claims

1. A system comprising:

at least one accelerometer configured to be positioned on a patient, the at least one accelerometer configured to generate acceleration data;

a processor in communication with the at least one accelerometer, the processor configured to;

receive acceleration data from the at least one accelerometer;

determine a plurality of components of the acceleration data;

calculate, for each of the plurality of components, a non-linear energy of the component by applying one or more non-linear operators to the component;

calculate a non-linear energy of the acceleration data by combining the non-linear energies of the plurality of components of the acceleration data;

perform a first comparison of the non-linear energy of the acceleration data to a first threshold;

determine, based on the first comparison, whether the nonlinear energy of the acceleration data indicates an event; and

if the non-linear energy indicates the event;

perform a second comparison of the acceleration data to a second threshold; and

determine, based on the second comparison, whether the event is a seizure; and

a warning device that activates an alarm in response to receiving an indication from the processor that the event is a seizure.

2. The system of claim 1, further comprising:

a database configured to maintain a record of one or more events.

3. The system of claim 1, wherein the at least one accelerometer includes a first accelerometer and a second accelerometer, wherein the first accelerometer and the second accelerometer are positioned at different locations on the patient.

4. The system of claim 1, wherein the plurality of components of the acceleration data include a dynamic activity of the acceleration data, and wherein the second threshold includes a dynamic activity threshold.

5. The system of claim 1, wherein the plurality of components of the acceleration data include frequency domain features of the acceleration data, and wherein the second threshold includes a frequency threshold.

6. The system of claim 1, wherein at least one of the non-linear operators applied to the acceleration data by the processor is a Lie Bracket operator.

7. The system of claim 6, wherein the acceleration data includes a first acceleration component having an x component, a second acceleration component having a y component, and a third acceleration component having a z component.

8. The system of claim 7, wherein the non-linear energy is determined by calculating a non-linear energy for each component of the first, second, and third acceleration components to provide three results and averaging the three results together to calculate the non-linear energy.

9. The system of claim 1, further comprising:

one or more physiological monitoring systems for collecting one or more physiological measurements of the patient, the one or more physiological monitoring systems that send the one or more physiological measurements to the processor, wherein the processor analyzes the one or more physiological measurements to determine whether the one or more physiological measurements support a determination that the event is a seizure.

10. The system of claim 9, wherein the one or more physiological monitoring systems includes an EKG monitoring system that collects EKG measurements from the patient.

11. The system of claim 10, wherein the EKG monitoring system analyzes the EKG measurements to determine whether the EKG measurements support a determination that the event is a seizure, wherein the determination is based on a use of a relative heart rate detection algorithm.

12. The system of claim 11, wherein the EKG monitoring system performs a noise analysis, wherein the EKG measurements are disregarded to determine whether the event is a seizure when the noise analysis determines that a level of noise in the EKG measurements is too high.

13. The system of claim 9, wherein the one or more physiological monitoring systems includes one or more of a respiration monitoring system, or a skin resistivity monitoring system.

14. An apparatus comprising:

at least one accelerometer configured to be positioned on a patient, the at least one accelerometer configured to generate acceleration data;

a processor in configured to communicate with the at least one accelerometer and configured to;

receive the acceleration data from the at least one accelerometer;

determine a first dimension, a second dimension and a third dimension of the acceleration data;

calculate a first non-linear energy of the first dimension of the acceleration data by applying a first non-linear operator to the first dimension of the acceleration data;

calculate a second non-linear energy of the second dimension of the acceleration data by applying a second non-linear operator to the second dimension of the acceleration data;

calculate a third non-linear energy of the third dimension of the acceleration data by applying a third non-linear operator to the third dimension of the acceleration data;

calculate a non-linear energy of the acceleration data by combining the first non-linear energy, the second non-linear energy, and the third non-linear energy of the acceleration data;

perform a first comparison of the non-linear energy of the acceleration data to a first threshold to determine whether the non-linear energy of the acceleration data indicates an event; and

if the non-linear energy indicates the event;

perform a second comparison of the acceleration data to a second threshold; and

determine, based on the second comparison, whether the event is a seizure; and

a warning device that activates an alarm in response to receiving an indication from the processor that the event is a seizure.

15. The apparatus of claim 14, further comprising a memory to store settings to indicate when the patient is engaged in different levels of activity, wherein the levels of activity include strenuous activity, normal activity, sedentary activity or sleeping activity.

16. A method, comprising:

receiving acceleration data at a processor from at least one accelerometer positioned on a patient;

determining a plurality of components of the acceleration data;

calculating, for each of the plurality of components, a non-linear energy of the components by applying a non-linear operator to the component at the processor;

calculating a non-linear energy of the acceleration data by combining the non-linear energies of the components of the acceleration data;

performing a first comparison of the non-linear energy of the acceleration data to a first threshold;

determining, based on the first comparison, whether the non-linear energy of the acceleration data indicates an event;

if the non-linear energy indicates the event;

performing a second comparison of the acceleration data to a second threshold;

determining, based on the second comparison, whether the event is a seizure; and

activating an alarm on a warning device in response to determining the event is a seizure.

17. The method of claim 16, wherein performing the second comparison of the acceleration data to the second threshold further comprises:

measuring frequency domain features of the acceleration data to determine a maximum frequency; and

comparing the maximum frequency to the second threshold.

18. The method of claim 16, further comprising:

maintaining a record in a database of one or more events.

19. The method of claim 16, wherein the plurality of components of the acceleration data include dynamic activity of the acceleration data and, wherein the second threshold includes a dynamic activity threshold.

20. The method of claim 16, wherein the plurality of components of the acceleration data include frequency domain features of the acceleration data, and wherein the second threshold includes a frequency threshold.

21. The method of claim 16, wherein the non-linear operator applied to the acceleration data by the processor is a Lie Bracket operator.

22. The method of claim 21, wherein application of the Lie Bracket operator to the acceleration data includes use of a second order Lie Bracket equation.

23. The method of claim 21, wherein the acceleration data received at the processor includes a first acceleration component having an x component, a second acceleration component having a y component, and a third acceleration component having a z component.

24. The method of claim 23, wherein calculating the non-linear energy further comprises calculating a non-linear energy for each component of the first, second, and third acceleration components to provide three results and averaging the three results to provide the non-linear energy.

25. The method of claim 16, further comprising sending an activation signal to a warning device in response to determining that the event is a seizure.

26. The method of claim 25, wherein the warning device is configured to electrically stimulate a cranial nerve or brain tissue in response to the activation signal.