Patient-specific template development for neurological event detection

• US 7,136,695 B2
• Filed: 10/12/2001
• Issued: 11/14/2006
• Est. Priority Date: 10/12/2001
• Status: Expired due to Term

• Current Assignee: Neuropace Inc.
• Original Assignee: Neuropace Inc.
• Inventors: Gibb, Barbara, Kishawi, Eyad, Esteller, Rosana, Echauz, Javier, Tcheng, Thomas K., Pless, Benjamin D.
• Primary Examiner(s): LAYNO, CARL HERNANDZ
• Application Number:
  US09/977,052
• Publication Number:
  US 20030074033A1
• Time in Patent Office: 1,859 Days
• Field of Search: 600/509, 600/544, 607/32, 607/45, 607/60, 128/903
• US Class Current: 600/544
• CPC Class Codes: A61B 5/369 Electroencephalography [EEG] devices for psychotechnics A61B5/16
• CPC Class Codes: A61B 5/372 Analysis of electroencephalograms
• CPC Class Codes: A61B 5/4094 Diagnosing or monitoring seizure diseases, e.g. epilepsy
• CPC Class Codes: A61B 5/7217 of noise originating from a therapeutic or surgical apparatus, e.g. from a pacemaker
• CPC Class Codes: G16H 20/30 relating to physical therapies or activities, e.g. physiotherapy, acupressure or exercising
• CPC Class Codes: G16H 40/60 for the operation of medical equipment or devices

Abstract

An epileptiform activity patient-specific template creation system permits a user to efficiently develop an optimized set of patient-specific parameters for epileptiform activity detection algorithms. The epileptiform activity patient template creation system is primarily directed for use with an implantable neurostimulator system having EEG storage capability, in conjunction with a computer software program operating within a computer workstation having a processor, disk storage and input/output facilities for storing, processing and displaying patient EEG signals. The implantable neurostimulator is operative to store records of EEG data when neurological events are detected, when it receives external commands to record, or at preset or random times. The computer workstation operates on stored and uploaded records of EEG data to derive the patient-specific templates.

Claims

1. A method for developing a patient-specific detection template from at least one record of electrographic signal data stored by an implantable device, the method comprising the steps of:

uploading the at least one record of electrographic signal data from the implantable device to an external apparatus;

processing the at least one record and an event annotation to generate at least one parameter; and

forming the patient-specific detection template from the at least one parameter; and

downloading the patient-specific detection template from the external apparatus to the implantable device, wherein the patient-specific detection template comprises a prediction template.

2. A method for developing a patient-specific detection template from at least one record of electrographic signal data stored by an implantable device, the method comprising the steps of:

uploading the at least one record of electrographic signal data from the implantable device to an external apparatus;

processing the at least one record and an event annotation to generate at least one parameter; and

forming the patient-specific detection template from the at least one parameter; and

downloading the patient-specific detection template from the external apparatus to the implantable device.

3. The method for developing a patient-specific detection template of claim 2, wherein forming the patient-specific detection template comprises creating a data structure including at least one data item representative of the at least one parameter.

4. The method for developing a patient-specific detection template of claim 2, further comprising the step of creating the event annotation by annotating at least one event in the at least one record.

5. The method for developing a patient-specific detection template of claim 4, wherein the step of creating the event annotation comprises the steps of:

viewing the at least one record of electrographic signal data; and

indicating a start time of the event with the event annotation.

6. The method for developing a patient-specific detection template of claim 5, wherein the step of creating the event annotation further comprises the step of indicating an end time of the event with the event annotation.

7. The method for developing a patient-specific detection template of claim 4, wherein the step of creating the event annotation comprises the steps of:

obtaining data representative of an event log containing an event entry; and

indicating a start time of the event with the event annotation by processing the event entry.

8. The method for developing a patient-specific detection template of claim 7, wherein the step of creating the event annotation further comprises the steps of:

viewing the at least one record of electrographic signal data; and

modifying the event annotation to refine the start time of the event.

9. The method for developing a patient-specific detection template of claim 2, further comprising the step of augmenting the at least one record of electrographic data with an artificial data record.

10. The method for developing a patient-specific detection template of claim 9, wherein the augmenting step comprises the steps of:

transforming a subset of the at least one record of electrographic data to create an additional artificial data record; and

adding the artificial data record to the at least one record prior to the developing step.

11. The method for developing a patient-specific detection template of claim 10, wherein the transforming step comprises adjusting a playback speed of the subset.

12. The method for developing a patient-specific detection template of claim 10, wherein the transforming step comprises adjusting an amplitude of the subset.

13. The method for developing a patient-specific detection template of claim 10, wherein the transforming step comprises adding a noise component to the subset.

14. The method for developing a patient-specific detection template of claim 13, wherein the subset represents baseline electrographic signal data, and wherein the noise component includes artificially generated noise, stereotypical environmental noise, or a noise record from another patient.

15. The method for developing a patient-specific detection template of claim 13, wherein the subset represents event-containing electrographic signal data, and wherein the noise component includes artificially generated in-band noise.

16. The method for developing a patient-specific detection template of claim 2, wherein the step of forming a detection template comprises the steps of:

selecting a set of device parameters; and

testing the set of device parameters on at least one test record.

17. The method for developing a patient-specific detection template of claim 16, wherein the set of device parameters comprises at least one detection parameter.

18. The method for developing a patient-specific detection template of claim 16, wherein the set of device parameters comprises at least one prediction parameter.

19. The method for developing a patient-specific detection template of claim 16, wherein the set of device parameters comprises at least one signal pre-processing parameter.

20. The method for developing a patient-specific detection template of claim 16, further comprising the step of choosing a detection algorithm.

21. The method for developing a patient-specific detection template of claim 16, further comprising the step of selecting the at least one test record as a subset of the at least one record of electrographic signal data.

22. The method for developing a patient-specific detection template of claim 16, wherein the step of selecting a set of device parameters is performed automatically.

23. The method for developing a patient-specific detection template of claim 16, wherein the step of testing the detection parameters comprises the steps of:

identifying a plurality of performance factors on the at least one test record;

combining the plurality of performance factors into a performance metric.

24. The method for developing a patient-specific detection template of claim 16, wherein the step of testing the detection parameters comprises the steps of:

measuring a detection latency;

identifying any correct positive detections on the at least one record;

identifying any false positive detections on the at least one record;

identifying any false negative detections on the at least one record;

combining the detection latency, the correct positive detections, the false positive detections, and the false negative detections into a performance metric.

25. The method for developing a patient-specific detection template of claim 24, further comprising the steps of:

identifying a false positive detection occurring before the event annotation;

changing the false positive detection into a predictive correct positive detection; and

calculating a negative detection latency between the predictive correct positive detection and the event annotation.

26. The method for developing a patient-specific detection template of claim 16, wherein the step of testing the detection parameters comprises the steps of:

identifying any point correct positive detections on the at least one test record;

measuring a detection latency;

identifying any block correct positive detections on the at least one record;

identifying any point false positive detections on the at least one record;

identifying any block false positive detections on the at least one record;

scaling the number of block false positives by a total block false positive duration;

identifying any block false negative detections on the at least one record;

combining the point correct positive detections, the detection latency, the block correct positive detections, the point false positive detections, the scaled block false positive detections, and the block false negative detections into a performance metric.

27. The method for developing a patient-specific detection template of claim 16, further comprising the step of repeating the selecting step and the testing step.

28. The method for developing a patient-specific detection template of claim 16, further comprising the step of saving the detection parameters.

29. The method for developing a patient-specific detection template of claim 2, wherein the step of forming a detection template comprises performing a greedy line search.

30. The method for developing a patient-specific detection template of claim 29, wherein performing a greedy line search comprises the steps of:

selecting a variable parameter from a working parameter set;

testing the working parameter set on at least one test record;

adjusting a value of the variable parameter;

repeating the testing and adjusting steps over a range of values of the variable parameter; and

identifying a desired value of the variable parameter.

31. The method for developing a patient-specific detection template of claim 30, further comprising the step of selecting the at least one test record as a subset of the at least one record of electrographic signal data.

32. The method for developing a patient-specific detection template of claim 30, further comprising the step of incorporating the desired value of the variable parameter into the working parameter set.

33. The method for developing a patient-specific detection template of claim 30, further comprising the steps of:

selecting an initial parameter set; and

deriving the working parameter set from the initial parameter set.

34. The method for developing a patient-specific detection template of claim 33, wherein selecting an initial parameter set comprises the steps of:

employing at least one heuristic algorithm to generate at least one initial parameter; and

forming an initial parameter set from the at least one initial parameter.

35. The method for developing a patient-specific detection template of claim 30, further comprising the step of selecting a termination criterion.

36. The method for developing a patient-specific detection template of claim 35, further comprising the step of comparing the working parameter set to a stored parameter set.

37. The method for developing a patient-specific detection template of claim 36, wherein the step of comparing the working parameter set to a stored parameter set comprises the steps of:

identifying a plurality of performance factors on the at least one test record;

combining the plurality of performance factors into a working performance metric; and

comparing the working performance metric to a stored performance metric corresponding to the stored parameter set.

38. The method for developing a patient-specific detection template of claim 37, further comprising the step of replacing the stored parameter set with the working parameter set if the working performance metric is better than the stored performance metric.

39. The method for developing a patient-specific detection template of claim 36, further comprising the step of determining whether the termination criterion is met.

40. The method for developing a patient-specific detection template of claim 35, further comprising the step of iteratively testing parameter sets to accomplish a multidimensional search by repeating the steps of selecting a parameter, testing the working parameter sets, repeating the testing and adjusting steps, and identifying a preferred parameter value.

41. The method for developing a patient-specific detection template of claim 40, wherein the step of iteratively testing parameter sets is performed repeatedly until the termination criterion is met.

42. The method for developing a patient-specific detection template of claim 41, wherein the termination criterion comprises a convergence bound.

43. The method for developing a patient-specific detection template of claim 41, wherein the termination criterion comprises a maximum number of iterations.

44. The method for developing a patient-specific detection template of claim 30, further comprising the step of converting the patient-specific detection template into a format recognizable by the implantable device.

45. The method for developing a patient-specific detection template of claim 2, further comprising step of evaluating the detection template before the downloading step.

46. The method for developing a patient-specific detection template of claim 45, wherein the evaluating step comprises determining whether the detection template is satisfactory by visual inspection of a simulation.

47. The method for developing a patient-specific detection template of claim 45, wherein the evaluating step comprises the step of checking the detection template against a plurality of records of electrographic data.

48. The method for developing a patient-specific detection template of claim 2, further comprising the step of manually refining the detection template before the downloading step.

49. The method for developing a patient-specific detection template of claim 2, wherein the at least one record comprises electrographic data representative of a plurality of channels, and wherein the method further comprises the step of identifying a channel of interest.

50. The method for developing a patient-specific detection template of claim 49, wherein the step of identifying a channel of interest comprises creating a processed channel from a plurality of channels in the record.

51. The method for developing a patient-specific detection template of claim 2, wherein the at least one record comprises a plurality of records, and further comprising the step of dividing the plurality of records into a plurality of sets of records representing different types of data.

52. The method for developing a patient-specific detection template of claim 51, wherein the plurality of sets of records comprises a first set of records and a second set of records, and wherein the first set of records represents event-containing electrographic signal data and the second set of records represents baseline electrographic signal data.

53. The method for developing a patient-specific detection template of claim 52, wherein the uploading step comprises the steps of:

receiving the first set of records from storage in the implantable device; and

receiving at least a portion of the second set of records from a real-time source of electrographic data.

54. The method for developing a patient-specific detection template of claim 53, wherein the real-time source of electrographic data is the implantable device.

55. The method for developing a patient-specific detection template of claim 2, wherein the uploading step comprises the step of causing the implantable device to transmit the at least one record over a wireless link to the external apparatus.

56. The method for developing a patient-specific detection template of claim 2, wherein the uploading step comprises the step of causing a database to transmit the at least one record to the external apparatus, wherein the at least one record originated in the implantable device.

57. The method for developing a patient-specific detection template of claim 2, wherein the uploading step comprises the step of causing a transceiver to obtain the at least one record from the implantable device over a wireless link and transmit the at least one record over a secondary link to the external apparatus.

58. The method for developing a patient-specific detection template of claim 57, wherein the secondary link comprises a telephone line.

59. The method for developing a patient-specific detection template of claim 57, wherein the secondary link comprises a computer network.

60. The method for developing a patient-specific detection template of claim 2, wherein the external apparatus is a programmer.

61. The method for developing a patient-specific detection template of claim 2, wherein the external apparatus is a database.

62. A system for developing a patient-specific template for identifying neurological activity in a human patient, comprising:

an implantable device having a control module and at least one electrode adapted to receive an electrical signal from the patient'"'"'s brain; and

an external apparatus capable of bi-directional communication with the implantable device;

wherein the template comprises an operational parameter and the external apparatus is adapted to transmit the operational parameter to the implantable device;

wherein the implantable device is adapted to record at least one record of the electrical signal in accordance with the operational parameter and transmit the record to the external apparatus; and

wherein the external apparatus is adapted to receive and perform an operation on the record.

63. The system for developing a patient-specific template of claim 62, wherein the operational parameter defines at least one specified recording start time.

64. The system for developing a patient-specific template of claim 63, wherein the operational parameter further specifies at least one recording duration.

65. The system for developing a patient-specific template of claim 63, wherein the operational parameter further specifies at least one recording end time.

66. The system for developing a patient-specific template of claim 63, wherein the specified recording start time is recurring.

67. The system for developing a patient-specific template of claim 62, wherein the electrical signal comprises a bipolar EEG signal received from two electrodes.

68. The system for developing a patient-specific template of claim 62, wherein the operational parameter indicates at least one random start time.

69. The system for developing a patient-specific template of claim 62, wherein the control module comprises an electronic circuit having a detection subsystem, and wherein the detection subsystem is programmable with a signal processing parameter.

70. The system for developing a patient-specific template of claim 69, wherein the operational parameter comprises the signal processing parameter.

71. The system for developing a patient-specific template of claim 70, wherein the external apparatus comprises a programmer.

72. The system for developing a patient-specific template of claim 71, wherein the programmer is programmed with event detection template development software operative to develop a template from the record.

73. The system for developing a patient-specific template of claim 72, wherein the signal processing parameter is generated by the event detection template development software.

74. The system for developing a patient-specific template of claim 69, wherein the operational parameter indicates an event to be detected by the detection subsystem.

75. The system for developing a patient-specific template of claim 74, wherein the operational parameter further indicates an action to be performed in response to the event.

76. The system for developing a patient-specific template of claim 75, wherein the action comprises storing a record of the electrical signal.

77. The system for developing a patient-specific template of claim 62, wherein the external apparatus comprises a display unit adapted to simultaneously display a plurality of records received from the implantable device.

78. The system for developing a patient-specific template of claim 77, wherein the external apparatus further comprises an input device allowing a user to annotate an event in a record of the plurality of records.

79. The system for developing a patient-specific template of claim 78, wherein the event comprises a clinical onset of a seizure.

80. The system for developing a patient-specific template of claim 78, wherein the event comprises an electrographic onset of a seizure.

81. The system for developing a patient-specific template of claim 78, wherein the event comprises an electrographic precursor to a seizure.

82. The system for developing a patient-specific template of claim 78, wherein the event comprises an electrographic event predictive of a seizure.

83. The system for developing a patient-specific template of claim 62, further comprising an initiating device capable of transmitting at least one code to the implantable device, wherein the implantable device is capable of receiving and acting in response to the code.

84. The system for developing a patient-specific template of claim 83, wherein the initiating device is capable of being operated by the patient.

85. The system for developing a patient-specific template of claim 83, wherein the initiating device comprises a magnet.

86. The system for developing a patient-specific template of claim 83, wherein the initiating device comprises an external electronic device.

87. The system for developing a patient-specific template of claim 83, wherein the initiating device is capable of transmitting a code representative of a command to store a timestamp of an event sensed by the patient or a code representative of a command to begin recording.