SENSORY-EVOKED POTENTIAL (SEP) CLASSIFICATION/DETECTION IN THE TIME DOMAIN

• US 20130211276A1
• Filed: 01/31/2013
• Published: 08/15/2013
• Est. Priority Date: 03/16/2009
• Status: Abandoned Application

• Current Assignee: Neurosky Incorporated
• Original Assignee: Neurosky Incorporated
• Inventors: Luo, An, Sullivan, Thomas J., Delorme, Arnaud
• Application Number:
  US13/756,403
• Publication Number:
  US 20130211276A1
• Time in Patent Office: Days
• US Class Current: 600/544
• CPC Class Codes: A61B 5/165 Evaluating the state of mind, e.g. depression, anxiety
• CPC Class Codes: A61B 5/18 for vehicle drivers or machine operators
• CPC Class Codes: A61B 5/372 Analysis of electroencephalograms
• CPC Class Codes: A61B 5/378 Visual stimuli
• CPC Class Codes: A61B 5/38 Acoustic or auditory stimuli
• CPC Class Codes: A61B 5/7203 for noise prevention, reduction or removal
• CPC Class Codes: A61B 5/7264 Classification of physiological signals or data, e.g. using neural networks, statistical classifiers, expert systems or fuzzy systems neural networks per se G06N3/00; expert systems per se G06N5/00
• CPC Class Codes: A61B 5/7267 involving training the classification device
• CPC Class Codes: G06F 3/015 Input arrangements based on nervous system activity detection, e.g. brain waves [EEG] detection, electromyograms [EMG] detection, electrodermal response detection
• CPC Class Codes: G16H 50/20 for computer-aided diagnosis, e.g. based on medical expert systems

Abstract

Techniques are disclosed for sensory-evoked potential (SEPs, e.g., visual-evoked potentials) signal detection/classification by synchronizing EEG to the repeated presentation of sensory stimuli in the time domain. In some embodiments, a system receives a plurality of EEG signal samples, generates a stimulus-locked EEG and determines whether the plurality of EEG signal samples are evoked in response to a pattern of stimulus. In some embodiments, no prior knowledge about the update pattern (such as the flashing frequency of a visual stimulus) of the stimulus and no prior knowledge about an individual user'"'"'s EEG pattern are required.

Claims

1-22. -22. (canceled)

23. A method for sensory-evoked potential signal classification, comprising:

receiving a signal that is used to control a pattern of stimulus;

generating the pattern of stimulus characterized by a frequency that updates randomly;

determining in real-time the frequency of the pattern of stimulus, the determined frequency of the pattern of stimulus being the current frequency of the pattern of stimulus;

receiving a plurality of electroencephalography signal samples from a current user evoked in response to the pattern of stimulus;

synchronizing the plurality of electroencephalography signal samples in a time domain without conversion to a frequency domain to the pattern of stimulus in the time domain;

generating a stimulus-locked electroencephalography signal sample using a processor; and

performing a determination using a time domain classifier in real-time of whether the plurality of electroencephalography signal samples are evoked in response to the pattern of stimulus using the processor,

wherein the determination of whether the plurality of electroencephalography signal samples are evoked in response to the pattern of stimulus is performed dynamically or adaptively in real-time in a manner such that the current user'"'"'s electroencephalography pattern determined in real-time is used, and the frequency of the pattern of stimulus determined in real-time is used.

24. The method recited in claim 23, wherein the current user'"'"'s electroencephalography response pattern determined before the current user'"'"'s current electroencephalography response pattern is not used in performing the determination of whether the plurality of electroencephalography signal samples are evoked in response to the pattern of stimulus, and a frequency of the pattern of stimulus determined before the current determined frequency of the pattern of stimulus is not used in the determination of whether the plurality of electroencephalography signal samples are evoked in response to the pattern of stimulus.

25. The method recited in claim 23, wherein the sensory-evoked potential signal classification is computed by:

calculating a mutual relationship between pairs of segments for the stimulus-locked electroencephalography signal sample;

determining a statistical measure of the mutual relationship for each stimulus; and

determining if a stimulus is being attended to by the current user according to the statistical measure of the mutual relationship for each stimulus user the time domain classifier.

26. The method recited in claim 23, wherein the sensory-evoked potential signal classification is performed by synchronizing electroencephalography signal samples to repeated presentation of sensory stimuli in the time domain.

27. The method recited in claim 23, wherein the sensory-evoked potential signal classification is performed by synchronizing electroencephalography signal samples to repeated presentation of sensory stimuli in the time domain, wherein the sensory-evoked potential signal classification includes visually evoked potentials.

28. The method recited in claim 23, further comprising:

detecting a plurality of electroencephalography signals.

29. The method recited in claim 23, further comprising:

recording the plurality of electroencephalography signal samples.

30. The method recited in claim 23, further comprising:

synchronizing the plurality of electroencephalography signal samples to the pattern of stimulus.

31. The method recited in claim 23, further comprising:

synchronizing each electroencephalography signal sample to an onset of the pattern of stimulus.

32. The method recited in claim 23, further comprising:

determining which stimulus is attended to by the current user.

33. The method recited in claim 23, further comprising:

generating a control signal based on the sensory-evoked potential signal classification.

34. The method recited in claim 23, wherein the mutual relationship includes correlation, mutual information, or covariance.

35. The method recited in claim 23, wherein the statistical measure includes a mean, a median value, or a sum.

36. A system for sensory-evoked potential signal classification, comprising:

a processor configured to;

receive a signal that is used to control a pattern of stimulus;

generate the pattern of stimulus characterized by a frequency that updates randomly;

determine in real-time the frequency of the pattern of stimulus, the determined frequency of the pattern of stimulus being the current determined frequency of the pattern of stimulus;

receive a plurality of electroencephalography signal samples from a current user evoked in response to the pattern of stimulus;

synchronize the plurality of electroencephalography signal samples in a time domain without conversion to a frequency domain to the pattern of stimulus in the time domain;

generate a stimulus-locked electroencephalography signal sample; and

perform a determination using a time domain classifier in real-time of whether the plurality of electroencephalography signal samples are evoked in response to the pattern of stimulus,

wherein the determination of whether the plurality of electroencephalography signal samples are evoked in response to the pattern of stimulus is performed dynamically or adaptively in real-time in a manner such that the current user'"'"'s electroencephalography pattern determined in real-time is used, and the frequency of the pattern of stimulus determined in real-time is used; and

a memory coupled to the processor and configured to provide the processor with instructions.

37. The system recited in claim 36, wherein the current user'"'"'s electroencephalography response pattern determined before the current user'"'"'s current electroencephalography response pattern is not used in performing the determination of whether the plurality of electroencephalography signal samples are evoked in response to the pattern of stimulus, and a frequency of the pattern of stimulus determined before the current determined frequency of the pattern of stimulus is not used in the determination of whether the plurality of electroencephalography signal samples are evoked in response to the pattern of stimulus.

38. The system recited in claim 36, wherein the sensory-evoked potential signal classification is computed by:

calculating a mutual relationship between pairs of segments for the stimulus-locked electroencephalography signal sample;

determining a statistical measure of the mutual relationship for each stimulus; and

determining if a stimulus is being attended to by the current user according to the statistical measure of the mutual relationship for each stimulus user the time domain classifier.

39. The system recited in claim 36, wherein sensory-evoked potential signal classification is performed by synchronizing electroencephalography signal samples to repeated presentation of sensory stimuli in the time domain, wherein the sensory-evoked potential signal classification includes visually evoked potentials.

40. A computer program product for sensory-evoked potential signal classification, the computer program product being embodied in a non-transitory computer readable storage medium and comprising computer instructions for:

receiving a signal that is used to control a pattern of stimulus;

generating the pattern of stimulus characterized by a frequency that updates randomly;

determining the frequency of the pattern of stimulus determined in real-time, the determined frequency of the pattern of stimulus being the current determined frequency of the pattern of stimulus;

receiving a plurality of electroencephalography signal samples from a current user evoked in response to the pattern of stimulus;

synchronizing the plurality of electroencephalography signal samples in a time domain without conversion to a frequency domain to the pattern of stimulus in the time domain;

generating a stimulus-locked electroencephalography signal sample using a processor; and

performing a determination using a time domain classifier in real-time of whether the plurality of electroencephalography signal samples are evoked in response to the pattern of stimulus using the processor,

wherein the determination of whether the plurality of electroencephalography signal samples are evoked in response to the pattern of stimulus is performed dynamically or adaptively in real-time in a manner such that the current user'"'"'s electroencephalography pattern determined in real-time is used, and the frequency of the pattern of stimulus determined in real-time is used.

41. The computer program product recited in claim 40, wherein the current user'"'"'s electroencephalography response pattern determined before the current user'"'"'s current electroencephalography response pattern is not used in performing the determination of whether the plurality of electroencephalography signal samples are evoked in response to the pattern of stimulus, and a frequency of the pattern of stimulus determined before the current determined frequency of the pattern of stimulus is not used in the determination of whether the plurality of electroencephalography signal samples are evoked in response to the pattern of stimulus.

42. The computer program product recited in claim 40, wherein the sensory-evoked potential signal classification is computed by:

calculating a mutual relationship between pairs of segments for the stimulus-locked electroencephalography signal sample;

determining a statistical measure of the mutual relationship for each stimulus; and

determining if a stimulus is being attended to by the current user according to the statistical measure of the mutual relationship for each stimulus user the time domain classifier.

43. The computer program product recited in claim 40, wherein sensory-evoked potential signal classification is performed by synchronizing electroencephalography signal samples to repeated presentation of sensory stimuli in the time domain, wherein the sensory-evoked potential signal classification includes visually evoked potentials.