Cerebral programming

• US 7,499,894 B2
• Filed: 09/15/2003
• Issued: 03/03/2009
• Est. Priority Date: 03/13/2001
• Status: Expired due to Fees

• Current Assignee: Benny Rousso, Goded Shahaf, Shimon Marom, Shlomo Ben Haim
• Original Assignee: Benny Rousso, Goded Shahaf, Shimon Marom, Shlomo Ben Haim
• Inventors: Shahaf, Goded, Rousso, Benny, Marom, Shimon, Ben Haim, Shlomo
• Primary Examiner(s): STARKS, WILBERT L
• Application Number:
  US10/662,987
• Publication Number:
  US 20040131998A1
• Time in Patent Office: 1,996 Days
• Field of Search: 706/23, 89/11.3, 102/355
• US Class Current: 706/23
• CPC Class Codes: A61N 1/36017 with leads or electrodes penetrating the skin
• CPC Class Codes: A61N 1/36021 for treatment of pain
• CPC Class Codes: A61N 1/36025 for treating a mental or cerebral condition
• CPC Class Codes: A61N 1/3603 Control systems
• CPC Class Codes: A61N 1/36071 Pain
• CPC Class Codes: A61N 1/36082 Cognitive or psychiatric applications, e.g. dementia or Alzheimer's disease
• CPC Class Codes: G06N 3/061 using biological neurons, e.g. biological neurons connected to an integrated circuit

Abstract

A method of training a biological neural network using a controller, comprising:

• • applying a cycle comprising:
    • stimulating a neural network by said controller applying at least an input signal to the network;
    • detecting an output response of the network by said controller; and
    • modifying said stimulation by said controller for at least a period of time if said response matches a desired at least approximate response; and
  • repeating said cycle of stimulation, detection and modification at least one more time until said neural network is trained to generate a desired output response for said input signal.

Claims

1. A method of training an in-vivo biological neural network in a living human using a controller, comprising:

applying a cycle comprising;

stimulating a neural network by said controller applying at least an input signal to the network;

detecting an output response of the network by said controller; and

modifying said stimulation by said controller for at least a period of time if said response matches a desired at least approximate response; and

repeating said cycle of stimulation, detection and modification at least one more time until said neural network is trained to generate a desired output response for said input signal.

2. A method according to claim 1, wherein said input signal is a specific signal and wherein stimulating comprises applying a less specific stimulation.

3. A method according to claim 1, wherein said modifying comprises modifying said input signal.

4. A method according to claim 1, wherein said modifying comprises stopping said stimulation.

5. A method according to claim 1, wherein said modifying is applied in a manner known to affect a stability of connections in said network.

6. A method according to claim 1, wherein said modifying is applied in a manner known to affect a learning behavior of said neurons of said network.

7. A method according to claim 1, wherein said stimulating comprises stimulating using a magnetic field stimulator.

8. A method according to claim 1, wherein said stimulating comprises stimulating using an electric current.

9. A method according to claim 1, wherein said modifying comprises applying a chemical agent.

10. A method according to claim 1, wherein said modifying comprises applying an electric field.

11. A method according to claim 1, comprising modifying said method by said controller in response to a response of said network to said method.

12. A method according to claim 1, wherein said desired output response is a particular defined response.

13. A method according to claim 1, wherein said desired output response is a random response pattern indicative of an unlearning by said network.

14. A method according to claim 1, wherein said desired output response is a spatial shift in a portion of said network that responses to said input, from a previous responding portion to a shifted responding portion.

15. A method according to claim 1, comprising changing said input pattern during a repetition cycle.

16. A method according to claim 1, comprising increasing a resolution of said input pattern between repetition cycles.

17. A method according to claim 1, comprising requiring an output response with a reduced range of acceptable values in subsequent repetition cycles.

18. A method according to claim 1, wherein repeating comprises repeating until an area of said network is organized to act as an input interface for said network for direct setting of network values not via a network'"'"'s standard inputs.

19. A method according to claim 1, wherein repeating comprises repeating until an area of said network is organized to act as an output interface for said network for direct receiving of network values not via a network'"'"'s standard outputs.

20. A method according to claim 1, comprising:

providing a complex output response pattern to inputs to be trained into said network;

dividing said response pattern into sub-patterns;

training said sub-patterns individually; and

linking said sub-patterns.

21. A method according to claim 1, comprising training a desired output response pattern to inputs as a whole into said network.

22. A method according to claim 1, wherein said network is a GI motor complex.

23. A method according to claim 1, wherein said network is a nervous plexus.

24. A method according to claim 1, wherein said network is comprised in a brain.

25. A method according to claim 24, wherein said controller replaces at least one natural learning related function of said brain.

26. A method according to claim 24, wherein said output is measured on said brain.

27. A method according to claim 24, wherein said output is measured as a response of said human.

28. A method according to claim 24, wherein said input is provided directly to said brain.

29. A method according to claim 24, wherein said stimulation is provided directly to said brain.

30. A method according to claim 24, wherein said stimulation is provided to said brain via natural senses.

31. A method according to claim 24, wherein said input is provided to said brain via natural senses.

32. A method according to claim 24, wherein modifying said stimulation comprising controlling the propagation of a signal inside said brain.

33. A method according to claim 24, wherein said method is used to rehabilitate an old function of said network.

34. A method according to claim 24, wherein said method is used to teach a new function to said network.

35. A method according to claim 24, wherein said method is used to teach new motor programs to said network.

36. A method according to claim 24, wherein said method is used to create a new pathway in said brain.

37. A method according to claim 24, wherein said method is used to create a new function area in said brain.

38. A method according to claim 24, wherein said method is used to remap a function from one brain area to another in said brain.

39. A method according to claim 24, wherein said method is used to erase a learned pattern from said brain.

40. A method according to claim 24, wherein said method is used to create a digital logic functioning area in said brain.

41. A method according to claim 24, wherein said method is used to train the brain in the use of an artificial organ.

42. A method according to claim 41, wherein said organ is a replacement organ.

43. A method according to claim 41, wherein said organ is a new organ not corresponding to a previous organ controlled by said brain.

44. A method according to claim 24, wherein said human is not conscious during the application of said method.

45. A method according to claim 24, wherein said human is not in control of learning processes imposed by said method.

46. A method according to claim 24, comprising not motivating said human in response to the output.

47. A method according to claim 24, wherein said human reports said output.

48. A method according to claim 24, wherein said human generates said input.

49. A method according to claim 24, comprising performing actions by said human to capture previously trained un-associated input-output response patterns.

50. A method according to claim 24, wherein said input is an input internal to said brain.

51. A method according to claim 24, wherein said output is an output internal to said brain.

52. A method according to claim 24, wherein said output comprises controlling an epilepsy attack.

53. A method according to claim 24, wherein said output comprises reducing a sensation of pain.

54. A method of training a biological neural network using a controller, comprising:

applying a cycle comprising;

stimulating a neural network by said controller applying at least an input signal, which is a localized signal, to the network, wherein said stimulating comprises applying a less localized stimulation;

detecting an output response of the network by said controller, and

modifying said stimulation by said controller for at least a period of time if said response matches a desired at least approximate response; and

repeating said cycle of stimulation, detection and modification at least one more time until said neural network is trained to generate a desired output response for said input signal.

55. A method according to claim 54, wherein said modifying comprises reducing said less localized stimulation.

56. A method according to claim 54, wherein said modifying comprises increasing said less localized stimulation.

57. A method according to claim 54, wherein said network is an in-vivo network in a living non-human animal.

58. A method according to claim 57, comprising:

providing a potential environmental contaminant;

repeating said method under at least two conditions of said contaminant; and

comparing a response of said network to said method to determine an effect of said contaminant on training of said animal.

59. A method according to claim 57, comprising:

selecting a network portion of said animal that is coupled to a sensing ability of said animal; and

training said network portion to output at least an indication of a sensing by said animal.

60. A method according to claim 57, wherein said input is a command that can be sensed by said animal and wherein said desired output is a behavioral response of said animal.

61. A method according to claim 60, wherein said input and said output are inter-related using a complex logic, including at least two logic steps.

62. An animal trained according to the method of claim 57.

63. A method of training an in vitro biological neural network using a controller, comprising:

applying a cycle comprising;

stimulating a neural network by said controller applying at least an input signal to the network;

detecting an output response of the network by said controller; and

modifying said stimulation by said controller for at least a period of time if said response matches a desired at least approximate response; and

repeating said cycle of stimulation, detection and modification at least one more time until said neural network is trained to generate a desired output response for said input signal.

64. A method according to claim 63, wherein said network is grown in a container.

65. A method according to claim 63, wherein said network is excised from a living body to a container.

66. A method according to claim 63, comprising:

providing a potential environmental contaminant;

repeating said method under at least two conditions of said contaminant; and

comparing a response of said network to said method to determine an effect of said contaminant on training of said network.