Brain imaging system and methods for direct prosthesis control

• US 8,583,565 B2
• Filed: 08/05/2010
• Issued: 11/12/2013
• Est. Priority Date: 08/03/2009
• Status: Expired due to Fees

• Current Assignee: University of Denver
• Original Assignee: University of Denver
• Inventors: Shoureshi, Rahmat A., Aasted, Christopher
• Primary Examiner(s): Gaffin, Jeffrey A
• Assistant Examiner(s): Traktovenko, Ilya
• Application Number:
  US12/850,964
• Publication Number:
  US 20110208675A1
• Time in Patent Office: 1,195 Days
• Field of Search: 706/2, 706/1, 706/62
• US Class Current: 706/2
• CPC Class Codes: G05B 13/0285 using neural networks and fuzzy logic
• CPC Class Codes: G06N 3/043 based on fuzzy logic, fuzzy membership or fuzzy inference, e.g. adaptive neuro-fuzzy inference systems [ANFIS]
• CPC Class Codes: G06N 3/08 Learning methods
• CPC Class Codes: G06N 5/048 Fuzzy inferencing

Abstract

Methods and systems for controlling a prosthesis using a brain imager that images a localized portion of the brain are provided according to one embodiment of the invention. For example, the brain imager can provide motor cortex activation data using near infrared imaging techniques and EEG techniques among others. EEG and near infrared signals can be correlated with brain activity related to limbic control and may be provided to a neural network, for example, a fuzzy neural network that maps brain activity data to limbic control data. The limbic control data may then be used to control a prosthetic limb. Other embodiments of the invention include fiber optics that provide light to and receive light from the surface of the scalp through hair.

Claims

1. A method for controlling a prosthesis, the method comprising:

receiving an IR brain activity signal that corresponds with at least a portion of the motor cortex at one or more IR photodiodes;

wherein the IR photodiodes transmit a plurality of inputs that correspond with the first portion of the brain;

receiving an EEG brain activity signal that corresponds with at least a portion of the motor cortex at one or more Electroencephalograph electrodes;

wherein the Electroencephalograph electrodes transmit a plurality of inputs that correspond with the second portion of the brain; and

wherein the said first portion of the brain is different from the said second portion of the brain;

mapping the IR brain activity signal and the EEG brain activity signal to an output signal that corresponds with a muscle group using a computer-based neural network that is trained to map the IR brain activity signal and the EEG brain activity signal to an output signal; and

coupling the output signal with the prosthesis, wherein the prosthesis is configured to respond to the output signal.

2. The method according to claim 1, wherein the neural network comprises a fuzzy neural network.

3. The method according to claim 1, wherein the method further comprises detecting action potentials from neurons within the portion of the motor cortex.

4. The method according to claim 1, wherein the method further comprises:

illuminating light on a portion of the brain using a light source; and

receiving a light signal at the scalp using a photodiode, wherein the detected light travels from the light source through a sub-portion of the brain and is detected at the photodiode.

5. The method according to claim 4, wherein the light source illuminates a portion of the brain with near infrared light.

6. The method according to claim 4, wherein the light source comprises at least one or more LEDs and one or more lasers.

7. A prosthesis control system, comprising:

one or more light sources configured to irradiate light into a first portion of the brain;

one or more IR photo diodes configured to detect a portion of the light transmitted into the first portion of the brain, wherein the detected light travels at least from the one or more light sources through a plurality of sub-portions of the brain and is detected at the plurality of IR photodiodes; and

wherein the IR photodiodes transmit a plurality of inputs that correspond with the first portion of the brain;

one or more Electroencephalograph electrodes configured to detect action potentials from neurons within a second portion of the brain; and

wherein the Electroencephalograph electrodes transmit a plurality of inputs that correspond with the second portion of the brain; and

wherein the said first portion of the brain is different from the said second portion of the brain; and

a controller coupled with the one or more IR photodiodes, the one or more Electroencephalograph electrodes, and the prosthesis;

wherein the controller is configured to receive a plurality of inputs from the plurality of IR photo diodes and the one or more Electroencephalograph electrodes;

wherein the controller is configured to;

determine the brain activity at a plurality of sub-portions of the first portion of the brain from the plurality of inputs from the plurality of IR photodiodes;

determine the brain activity at a plurality of sub-portions of the second portion of the brain from the plurality of inputs from the Electroencephalograph electrodes; and

determine a plurality of limbic control signals by mapping the brain activity within the first portion and the second portion of the brain with an output signal that corresponds with a muscle group.

8. The prosthesis control system according to claim 7 further comprising a prosthesis coupled with the controller and wherein the controller is further configured to send the plurality of limbic control signals to the prosthesis.

9. The prosthesis control system according to claim 7, wherein the determining the brain activity at a plurality of sub-portions of the first portion of the brain further comprises:

determine the relative concentration of oxy-hemoglobin within the first portion of the brain from the plurality of IR photodiodes inputs;

determine the relative concentration of hemoglobin within the first portion of the brain from the plurality of IR photodiodes inputs; and

determine the brain activity at a plurality of sub-portions of the first portion of the brain from the relative concentrations of oxy-hemoglobin and hemoglobin.

10. The prosthesis control system according to claim 7, wherein the first and second portions of the brain comprise the same portion of the brain.

11. The prosthesis control system according to claim 7, wherein the controller comprises a neural network inference engine configured to determine the plurality of limbic control signals from the brain activity within the first portion of the brain.

12. The prosthesis control system according to claim 7, wherein the controller comprises a neural network.

13. The prosthesis control system according to claim 7, wherein the controller comprises a fuzzy neural network.

14. The prosthesis control system according to claim 7 comprising a muscle activity sensor, wherein the controller utilizes brain activity at the plurality of sub-portions of the first portion of the brain as feedforward signals and muscle activation as recorded by the muscle activity sensor as feedback signals.

15. A method for training a prosthesis system, wherein the prosthesis system comprises a neural network, an IR imaging system, an EEG system, and a prosthesis, the training utilizing a muscle activity detector, the method comprising:

detecting IR brain activity with the IR imaging system having one or more IR photodiodes;

wherein the IR photodiodes transmit a plurality of inputs that correspond with the first portion of the brain;

detecting EEG brain activity with the EEG system having one or more Electroencephalograph electrodes;

wherein the Electroencephalograph electrodes transmit a plurality of inputs that correspond with the second portion of the brain; and

wherein the said first portion of the brain is different from the said second portion of the brain;

receiving muscular response data from the muscle activity detector;

wherein the muscular response data is correlated with the IR and EEG brain activity using the neural network; and

training the neural network to produce the muscular response data from the brain activity data.

16. The method according to claim 15, wherein the neural network is a fuzzy neural network.

17. The method according to claim 15, wherein the muscle activity detector comprises an electromyograph.