Using a genetic algorithm to fit a cochlear implant system to a patient

• US 20050107845A1
• Filed: 10/14/2004
• Published: 05/19/2005
• Est. Priority Date: 03/11/2003
• Status: Active Grant

• Current Assignee: Cochlear Limited
• Original Assignee: Cochlear Limited
• Inventors: Van Den Honert, Christopher, Wakefield, Gregory H., Parker, John, Lineaweaver, Sean, Parkinson, Wendy, Patrick, Jim, Parkinson, Aaron
• Granted Patent: US 8,301,259 B2
• Time in Patent Office: Days
• US Class Current: 607/57
• CPC Class Codes: A61N 1/0541 Cochlear electrodes
• CPC Class Codes: A61N 1/36039 fitting procedures

Abstract

Apparatus and method for at least partially fitting a cochlear implant system to a patient is described, comprising: executing a genetic algorithm to select values for a subset of one or more parameters selected from a plurality of parameters for which values are to be selected to fit the implant, wherein said genetic algorithm is operable to generate one or more successive generations of values for said parameter subset; and determining, based on patient feedback, said values of said values for said parameter subset in each of said one or more successive generations.

Claims

1. A method for at least partially fitting a medical implant system to a patient comprising:

executing a genetic algorithm to select values for a subset of one or more parameters selected from a plurality of parameters for which values are to be selected to fit said medical implant, wherein said genetic algorithm is operable to generate one or more successive generations of values for said parameter subset; and

determining, based on patient feedback, said values of said values for said parameter subset in each of said one or more successive generations.

2. The method of claim 1, wherein said one or more parameters interact non-linearly.

3. The method of claim 1, wherein said one or more parameters interact non-monotonically.

4. The method of claim 1, wherein said subset of one or more parameters comprises at least one of a group consisting of:

a speech strategy implemented in said medical implant system;

a quantity of channels represented in said medical implant system;

which intracochlear and/or extracochlear electrodes are to be associated with each respective channel;

a pulse repetition rate for each respective channel;

a pulse width for each respective channel;

a number of spectral maxima periodically chosen for representation;

a mapping of sound pressure to stimulus current for each said respective channels;

front end filtering of audio from a microphone;

automatic gain control threshold;

compression ratio; and

attack and release times.

5. The method of claim 4, wherein said mapping of sound pressure to stimulus current for each channel comprises:

one or more threshold levels;

one or more comfort levels; and

one or more compression curves.

6. The method of claim 1, wherein said medical implant system provides electrical and mechanical stimulation, and wherein said subset of at least one parameter comprises:

at least one loudness parameter.

7. The method of claim 6, wherein said at least one loudness parameter comprises:

one or more long term loudness balance parameters.

8. The method of claim 6, wherein said at least one loudness parameter comprises:

one or more short term gain manipulations.

9. The method of claim 6, wherein said one or more short term gain manipulations comprises one or more signal-dependent gain adjustments.

10. The method of claim 9, wherein said one or more signal-dependent gain adjustments comprises one or more of a group consisting of:

adjustments to minimize cross-modal masking; and

adjustments to emphasize speech features.

11. The method of claim 10, wherein said speech features comprise one of frication and voicing.

12. The method of claim 1, wherein said medical implant system provides electrical and mechanical stimulation, and wherein said subset of at least one parameter comprises:

one or more frequency domain parameters.

13. The method of claim 12, wherein said one or more frequency domain parameters comprises:

frequency boundaries allocated to electrical and mechanical stimulation.

14. The method of claim 12, wherein said one or more frequency domain parameters comprises:

slopes of filtering at the boundaries of each stimulation signal.

15. The method of claim 12, wherein said one or more frequency domain parameters comprises:

allocation of frequency subbands in each domain.

16. The method of claim 1, wherein said cochlear implant system provides electrical and mechanical stimulation, and wherein said subset of at least one parameter comprises:

one or more time domain parameters.

17. The method of claim 16, wherein said one or more time domain parameters comprises:

adjusting electrical periodicity of pulse timing to be synchronized with mechanical signal fluctuations.

18. The method of claim 16, wherein said one or more time domain parameters comprises:

adjusting delays in electrical stimulus to compensate for missing propagation delays of various middle ear and inner ear pathways.

19. The method of claim 1, wherein said medical implant system provides electrical and mechanical stimulation, and wherein said subset of at least one parameter comprises:

one or more binaural parameters.

20. The method of claim 19, wherein said one or more binaural parameters comprises:

adjusting stimulus timing to present interaural timing cures.

21. The method of claim 19, wherein said one or more binaural parameters comprises:

adjustment of stimulus timing to suppress echo.

22. The method of claim 19, wherein said one or more binaural parameters comprises:

adjustment of stimulus timing to improve localization.

23. The method of claim 19, wherein said one or more binaural parameters comprises:

adjustment of stimulus timing to improve sound source segregation.

24. The method of claim 1, further comprising:

selecting values for parameters not included in said subset of at least one parameter by a method that does not use a genetic algorithm.

25. The method of claim 1, wherein said patient feedback comprises:

selecting, based on patient input, a specified quantity of values to survive in each generation from which said values for said parameter subset are determined for the next generation.

26. The method of claim 25, wherein values of said subset of parameters are represented by a string of bits.

27. The method of claim 26, wherein different sub-strings of bits in said string each represent values of respective parameters.

28. An apparatus for at least partially fitting a medical implant system to a patient comprising:

means for executing a genetic algorithm to select values for a subset of one or more parameters selected from a plurality of parameters for which values are to be selected to fit the implant, wherein said genetic algorithm is operable to generate one or more successive generations of values for said parameter subset; and

means for determining, based on patient feedback, said values of said values for said parameter subset in each of said one or more successive generations.

29. The method of claim 28, wherein said one or more parameters interact non-linearly.

30. The method of claim 28, wherein said one or more parameters interact non-monotonically.

31. The method of claim 28, wherein said subset of one or more parameters comprises at least one of a group consisting of:

a speech strategy implemented in said medical implant system;

a quantity of channels represented in said medical implant system;

which intracochlear and/or extracochlear electrodes are to be associated with each channel;

a pulse repetition rate for each channel;

a pulse width for each channel;

a number of spectral maxima periodically chosen for representation;

a mapping of sound pressure to stimulus current for each said channel;

front end filtering of the audio from the microphone;

automatic gain control threshold;

compression ratio; and

attack and release times.

32. The method of claim 31, wherein said mapping of sound pressure to stimulus current for each channel comprises:

one or more threshold levels;

one or more comfort levels; and

one or more compression curves.

33. The method of claim 28, wherein said medical implant system provides electrical and mechanical stimulation, and wherein said subset of at least one parameter comprises:

at least one loudness parameter.

34. The method of claim 33, wherein said at least one loudness parameter comprises:

one or more long term loudness balance parameters.

35. The method of claim 33, wherein said at least one loudness parameter comprises:

one or more short term gain manipulations.

36. The method of claim 33, wherein said one or more short term gain manipulations comprises one or more signal-dependent gain adjustments.

37. The method of claim 36, wherein said one or more signal-dependent gain adjustments comprises one or more of a group consisting of:

adjustments to minimize cross-modal masking; and

adjustments to emphasize speech features.

38. The method of claim 37, wherein said speech features comprise either frication or voicing.

39. The method of claim 28, wherein said medical implant system provides electrical and mechanical stimulation, and wherein said subset of at least one parameter comprises:

one or more frequency domain parameters.

40. The method of claim 39, wherein said one or more frequency domain parameters comprises:

frequency boundaries allocated to electrical and mechanical stimulation.

41. The method of claim 39, wherein said one or more frequency domain parameters comprises:

slopes of filtering at the boundaries of each stimulation signal.

42. The method of claim 39, wherein said one or more frequency domain parameters comprises:

allocation of frequency subbands in each domain.

43. The method of claim 28, wherein said medical implant system provides electrical and mechanical stimulation, and wherein said subset of at least one parameter comprises:

one or more time domain parameters.

44. The method of claim 43, wherein said one or more time domain parameters comprises:

adjusting electrical periodicity of pulse timing to be synchronized with mechanical signal fluctuations.

45. The method of claim 43, wherein said one or more time domain parameters comprises:

adjusting delays in the electrical stimulus to compensate for missing propagation delays of various middle ear and inner ear pathways.

46. The method of claim 28, wherein said cochlear implant system provides electrical and mechanical stimulation, and wherein said subset of at least one parameter comprises:

one or more binaural parameters.

47. The method of claim 46, wherein said one or more binaural parameters comprises:

adjusting stimulus timing to present interaural timing cures.

48. The method of claim 46, wherein said one or more binaural parameters comprises:

adjustment of stimulus timing to suppress echo.

49. The method of claim 46, wherein said one or more binaural parameters comprises:

adjustment of stimulus timing to improve localization.

50. The method of claim 46, wherein said one or more binaural parameters comprises:

adjustment of stimulus timing to improve sound source segregation.

51. The method of claim 28, further comprising:

selecting values for parameters not included in said subset of at least one parameter by a method that does not use a genetic algorithm.

52. The method of claim 28, wherein said patient feedback comprises:

selecting, based on patient input, a specified quantity of values to survive in each generation from which said values for said parameter subset are determined for the next generation.

53. The method of claim 52, wherein the values of said subset of parameters are represented by a string of bits.

54. The method of claim 53, wherein different sub-strings of bits in said string each represent values of respective parameters.

55. A method for at least partially fitting a medical implant system to a patient comprising:

executing a genetic algorithm to select values for a subset of one or more parameters selected from a plurality of parameters for which values are to be selected to fit said medical implant, wherein said genetic algorithm is operable to generate one or more successive generations of values for said parameter subset; and

determining, based on patient feedback, said values of said values for said parameter subset in each of said one or more successive generations,

wherein said medical implant system provides electrical and mechanical stimulation.

56. The method of claim 55, wherein said one or more parameters interact non-linearly.

57. The method of claim 55, wherein said one or more parameters interact non-monotonically.

58. The method of claim 55, wherein said subset of one or more parameters comprises:

a speech strategy implemented in said medical implant system.

59. The method of claim 55, wherein said subset of one or more parameters comprises:

a quantity of channels represented in said medical implant system.

60. The method of claim 55, wherein said subset of one or more parameters comprises:

which intracochlear and/or extracochlear electrodes are to be associated with each respective channel.

61. The method of claim 55, wherein said subset of one or more parameters comprises:

a pulse repetition rate for each respective channel.

62. The method of claim 55, wherein said subset of one or more parameters comprises:

a pulse width for each respective channel.

63. The method of claim 55, wherein said subset of one or more parameters comprises:

a number of spectral maxima periodically chosen for representation.

64. The method of claim 55, wherein said subset of one or more parameters comprises:

a mapping of sound pressure to stimulus current for each said respective channels.

65. The method of claim 55, wherein said subset of one or more parameters comprises:

front end filtering of audio from a microphone.

66. The method of claim 55, wherein said subset of one or more parameters comprises:

automatic gain control threshold.

67. The method of claim 55, wherein said subset of one or more parameters comprises:

compression ratio.

68. The method of claim 55, wherein said subset of one or more parameters comprises:

attack and release times.

69. The method of claim 55, wherein said mapping of sound pressure to stimulus current for each channel comprises:

one or more threshold levels;

one or more comfort levels; and

one or more compression curves.

70. The method of claim 55, wherein said subset of at least one parameter comprises:

at least one loudness parameter.

71. The method of claim 70, wherein said at least one loudness parameter comprises:

one or more long term loudness balance parameters.

72. The method of claim 70, wherein said at least one loudness parameter comprises:

one or more short term gain manipulations.

73. The method of claim 72, wherein said one or more short term gain manipulations comprises

one or more signal-dependent gain adjustments.

74. The method of claim 73, wherein said one or more signal-dependent gain adjustments comprises one or more of a group consisting of:

adjustments to minimize cross-modal masking; and

adjustments to emphasize speech features.

75. The method of claim 74, wherein said speech features comprise one of frication and voicing.

76. The method of claim 55, wherein said medical implant system provides electrical and mechanical stimulation, and wherein said subset of at least one parameter comprises:

one or more frequency domain parameters.

77. The method of claim 76, wherein said one or more frequency domain parameters comprises:

frequency boundaries allocated to electrical and mechanical stimulation.

78. The method of claim 76, wherein said one or more frequency domain parameters comprises:

slopes of filtering at the boundaries of each stimulation signal.

79. The method of claim 76, wherein said one or more frequency domain parameters comprises:

allocation of frequency subbands in each domain.

80. The method of claim 55, wherein said cochlear implant system provides electrical and mechanical stimulation, and wherein said subset of at least one parameter comprises:

one or more time domain parameters.

81. The method of claim 80, wherein said one or more time domain parameters comprises:

adjusting electrical periodicity of pulse timing to be synchronized with mechanical signal fluctuations.

82. The method of claim 80, wherein said one or more time domain parameters comprises:

adjusting delays in electrical stimulus to compensate for missing propagation delays of various middle ear and inner ear pathways.

83. The method of claim 55, wherein said medical implant system provides electrical and mechanical stimulation, and wherein said subset of at least one parameter comprises:

one or more binaural parameters.

84. The method of claim 83, wherein said one or more binaural parameters comprises:

adjusting stimulus timing to present interaural timing cures.

85. The method of claim 83, wherein said one or more binaural parameters comprises:

adjustment of stimulus timing to suppress echo.

86. The method of claim 83, wherein said one or more binaural parameters comprises:

adjustment of stimulus timing to improve localization.

87. The method of claim 83, wherein said one or more binaural parameters comprises:

adjustment of stimulus timing to improve sound source segregation.

88. The method of claim 55, further comprising:

selecting values for parameters not included in said subset of at least one parameter by a method that does not use a genetic algorithm.

89. The method of claim 55, wherein said patient feedback comprises:

selecting, based on patient input, a specified quantity of values to survive in each generation from which said values for said parameter subset are determined for the next generation.

90. The method of claim 89, wherein values of said subset of parameters are represented by a string of bits.

91. The method of claim 90, wherein different sub-strings of bits in said string each represent values of respective parameters.